Synaptic m1A remodeling in the mPFC promotes fear extinction via Pten regulation

Synaptic m1A remodeling in the mPFC promotes fear extinction via Pten regulation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Synaptic m1A remodeling in the mPFC promotes fear extinction via Pten regulation Xiang Li, Yi Zhang, Jiazhi Jiang, Ziwei Pi, Tongyu Chen, Zhipeng Xu, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7703002/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Post-traumatic stress disorder (PTSD) is marked by intrusive fear memories and impaired extinction learning, yet the underlying molecular mechanism at synapses remains elusive. Here, we uncover a previously unrecognized epitranscriptomic mechanism linking RNA modification to adaptive memory. We show that extinction learning increases synaptic TRMT6/TRMT61A levels, driving the redistribution of m¹A toward activity-relevant transcripts. Explicitly, using subcellular fractionation combined with m¹A RIP-seq, we identify a synapse-specific m¹A pattern, distinct from nuclear and whole mPFC profiles, marked by reduced 5’ UTR and enriched coding region and 3’ UTR methylation during extinction learning. Moreover, TRMT61A-mediated m¹A deposition in the 3’ UTR of Pten mRNA within synaptic compartments enhances transcript stability and thereby increases PTEN protein levels. RNA pull-down coupled with mass spectrometry identified SRSF1 as a synapse-enriched, extinction-responsive RNA binding protein (RBP) that specifically binds m¹A and mediates downstream post-transcriptional regulation. Notably, TRMT61A-mediated m¹A modification and its recognition by SRSF1 converge on Pten regulation to drive synaptic remodeling and facilitate extinction memory. Site-specific disrupting Pten m¹A site impairs dendritic structure and blocks extinction retrieval, completing a mechanistic loop from RNA mark to behavior. Our findings highlight m¹A as a dynamic regulator of synaptic plasticity and identify a novel molecular circuit that may be targeted for PTSD therapeutics. Biological sciences/Molecular biology Biological sciences/Neuroscience Health sciences/Diseases/Psychiatric disorders Biological sciences/Genetics Biological sciences/Psychology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Post-traumatic stress disorder (PTSD) is a trauma-related psychiatric condition characterized by the persistence of intrusive fear memories and impaired regulation of threat responses, leading to significant psychological and social dysfunction (American Psychiatric Association, DSM-5). While approximately 70% of individuals experience traumatic events during their lifetime[ 1 ], only a subset (~ 5.6%) develop PTSD[ 2 ], with lifetime prevalence estimates reaching up to 12.9% in certain populations[ 3 ], suggesting that individual differences in fear recovery mechanisms shape resilience or vulnerability. Among these, fear extinction, a form of inhibitory learning that enables the suppression of threat responses when danger is no longer present, has emerged as a core adaptive process disrupted in PTSD[ 4 ]. Accordingly, extinction-based psychotherapies, such as exposure therapy, represent the most effective interventions for PTSD[ 5 ]. However, therapeutic efficacy remains limited: remission is achieved in only approximately 50% of individuals with combat-related PTSD, and relapse rates remain high[ 6 ]. These limitations underscore a critical need to elucidate the molecular mechanisms underlying extinction learning and their dysfunction in PTSD. At the neural systems level, the medial prefrontal cortex (mPFC) plays a critical role in fear extinction by consolidating and retrieving fear extinction memory by inhibiting conditioned fear responses[ 7 ]. At the cellular level, extinction learning is supported by synaptic plasticity, the activity-dependent remodeling of synapses that enables durable behavioral adaptation[ 5 ]. Given the spatial segregation between synapses and the nucleus, local RNA regulation is crucial for stimulus-evoked protein synthesis within synaptic compartments[ 8 , 9 ]. Despite this, the molecular mechanisms governing RNA stability and translation at synapses during extinction learning remain incompletely understood. Epitranscriptomic regulation has recently emerged as an important mechanism for activity- and context-dependent gene expression in the brain[ 10 ]. While N⁶-methyladenosine (m⁶A) has been intensively characterized in the brain, recent work showed that synapse-enriched m⁶A-modified Malat1 interacts with a novel m⁶A reader at synapses[ 11 ], whereas the roles of other RNA modifications remain poorly understood. Among these modifications, N¹-methyladenosine (m¹A), a reversible modification installed by the TRMT6-TRMT61A methyltransferase complex and removed by ALKBH3, has been identified across diverse RNA species[ 12 – 16 ]. m¹A deposition at the 5’ untranslated region (UTR) enhances cap-independent translation initiation, whereas its presence in coding sequences(CDS) may impede elongation due to its positive charge[ 13 , 17 , 18 ]. Notably, neuronal contexts exhibit a redistribution of m¹A toward the 3’ UTR under stress conditions[ 19 , 20 ], suggesting a cell-type- and activity-specific reprogramming of m 1 A topology, though its functional role in synaptic compartments and behavior remains undefined. Here we examine the role and distribution of m¹A in extinction learning. We found that m¹A was dynamically regulated in extinction-trained mice, while knockdown of TRMT61A abolished this regulation and impaired extinction memory. Subcellular profiling revealed synapse-specific remodeling of m¹A, nuclear m¹A retained 5’ UTR enrichment regardless of condition, mirroring whole-tissue patterns; extinction triggered a synapse-specific shift from 5’ UTR toward CDS and 3’ UTR peaks. Among the modified transcripts, Pten emerged as a key synaptic target, regulated through TRMT61A-mediated m¹A and its recognition by SRSF1. Perturbation of this pathway disrupted dendritic architecture and impaired extinction memory, underscoring its behavioral significance. Together, our study uncovers an m¹A-centered regulatory pathway that links synaptic plasticity to extinction memory and provides new insight into the molecular basis of extinction memory formation. Method Plasma collection and m¹A quantification Peripheral blood samples were obtained from PTSD patients and healthy controls, matched for age and sex, with informed consent and IRB approval. (Zhongnan Hospital, Wuhan University, Kelun-2022174). Plasma was separated by centrifugation (1000g, 10min, 4°C) and stored at -80°C until analysis. m¹A levels were quantified using a commercial ELISA kit (Cell Biolabs, MET-5099) and performed according to the manufacturer’s instructions. Animals Male C57BL/6J mice (9–12 weeks) were housed under a 12-hour light/dark cycle with food and water ad libitum. All animal procedures were approved by the Animal Ethics Committee of Wuhan University (Approval No. ZN2023196). Cell culture Primary cortical neurons were isolated from E16 mouse embryos and cultured in Neurobasal medium (Gibco) supplemented with 1% GlutaMAX (Gibco), 2% B-27 (Gibco), and 1% penicillin/streptomycin (Servicebio, China). N2a and HEK293T cells were maintained in high-glucose DMEM ((Servicebio, China) with 10% FBS and 1% penicillin/streptomycin All kinds of cells were cultured at 37°C with 5% CO 2 . Cloning and lentiviral production CIRTS-PIN-Calm3 plasmid was obtained from the Timothy W. Bredy laboratory (The University of Queensland), with the cloning process detailed in previous publication[ 21 ]. ALKBH3 coding sequence was cloned into the CIRTS backbone to generate CIRTS-ALKBH3. shRNA constructs targeting TRMT61A or scrambled control (Table 1) were cloned into FG12H1 vector. Lentiviruses were produced in HEK293T cells using PEI-mediated transfection and ultracentrifugation. Viral titers > 1 × 10^8 IU/mL were used. Stereotactic viral injection Mice were anesthetized with 1% pentobarbital (50 mg/kg, i.p.) and secured in a stereotaxic frame (RWD Life Science, Shenzhen, China). A heating-pad maintained body temperature during the procedure. Lentivirus (500 µL per side, 75 nL/min) was bilaterally injected into the ILPFC (2.00 mm anterior to bregma, 0.20 mm lateral to midline, and 2.75 mm ventral to dura) using a glass pipette and microinjection pump (RWD Life Science). The pipette remained in place for 10 minutes to prevent viral spread before the brain wound was sutured. Mice were returned to home cages after awakening and allowed at least one week for recovery and stable viral expression. Behavioral assays (fear conditioning, extinction and open field) Two contexts (A and B) were used for fear conditioning and extinction behavioral testing. Conditioning chambers had transparent and stainless-steel walls with metal grid floors. In context B, the floor was covered with white plastic and lit by LED lights to reduce context generalization. Olfactory cues were also used to distinguish the contexts, with lemon scent applied in context A and vinegar in context B. Cameras recorded freezing behavior using Freezeframe software. Fear conditioning (FC) in context A included a 120 s pre-conditioning period, followed by three 120 s tones (80 dB, 16 kHz) paired with a 1 s, 0.6 mA foot shock. Mice were grouped based on freezing levels during the final tone and randomly assigned to different groups, ensuring comparable baseline freezing across groups. Extinction (EXT) in context B involved 30 non-reinforced 120 s tone presentations, while Retention Control (RC) mice were exposed to context B without tones. After 24 h, freezing was tested in context B during three 120 s tones, and memory was calculated as the percentage of freezing time. Open-field testing (OFT) was performed as described[ 22 ], with a shortened session (5 min). Tissue collection For tissue collection, fear acquisition was performed as in the behavioral tests. After 24 hours of FC, mice were exposed to 60 cycles of tones (EXT group) or silence (RC group). Tissues were collected immediately afterward. Synaptosome preparation mPFC tissue was homogenized and synaptosomes were isolated using Percoll gradient centrifugation as previously described[ 23 ]. Synaptosome purity was verified by Western blot. RNA isolation and RT-qPCR RNA was extracted using RNAiso Plus (Takara) and reverse-transcribed using PrimeScript Reverse Transcription Kit (Vazyme, China). Quantitative PCR was performed on a RotorGeneQ (Qiagen) cycler with SYBR Green Master Mix (Vazyme, China) and primers for target genes and PGK1 as a control (Supplementary Table). Transcript levels were normalized to PGK1 mRNA via the ΔΔCT method. Each PCR reaction was performed in duplicate for each sample and repeated at least twice. The primers used for RT-qPCR are listed in Table 1. Low input m1A-RIP seq The experiment was performed as previously described[ 11 ], with minor modifications. Briefly, Protein G magnetic beads (Thermo Fisher) were incubated with m¹A-specific antibody (MBL, China) in m¹A binding buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 2 mM EDTA, 1 mM DTT, RNase inhibitor) for 1 h at room temperature. Total RNA was fragmented with 10 mM ZnCl₂ in 10 mM Tris-HCl at 94°C for 1 min. immediately quenched on ice with 0.5 M EDTA, and incubated with antibody-coupled beads overnight at 4°C. Beads were sequentially washed with m¹A buffer, low-salt, high-salt, and TET buffers. m¹A-enriched RNA was eluted at 42°C using DTT-containing elution buffer and purified with RNA Clean & Concentrator Kit (Zymo). Libraries were constructed with SMARTer Stranded Pico Input RNA Kit v2 (Takara) and sequenced on an Illumina platform (PE150). RNA pull-down and mass spectrometry Biotinylated RNA probes containing m¹A or unmodified adenosine were synthesized (5’ Biotin-CCUGGGGG/m1A/AGACCCAGC-3’ and 5’ Biotin-CCUGGGGGAAGACCCAGC-3’) (TSINGKE, China). Nucleus-enriched and synaptosomal lysates from ILPFC of RC or EXT-trained mice were prepared in lysis buffer (10 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 10 mM Tris-HCl, pH 7.5, 0.5 mM DTT) with protease inhibitors (Roche). After preclearing with streptavidin beads (Thermo Fisher, #88817), lysates were incubated with RNA probes (2 µg) and captured on streptavidin beads pre-blocked with BSA and tRNA. Bound proteins were resolved by SDS-PAGE (10% Bis-Tris, Invitrogen), silver-stained (Solarbio), digested with trypsin, and identified by LC-MS/MS (Weizmann Institute MS Core). Two independent replicates were performed. Protein extraction and western blotting Tissues or subcellular fractions were lysed in RIPA buffer (Servicebio) containing protease inhibitors. Protein concentration was measured by BCA assay (Beyotime), and lysates were denatured at 95°C for 10 min. Proteins were separated by SDS-PAGE and transferred to PVDF membranes (Millipore). After blocking, membranes were probed with primary antibodies against TRMT61A, TRMT6, PSD95, SYN1, PTEN, GAPDH, ACTB, and H3, followed by HRP-conjugated secondary antibodies (Table 1). Chemiluminescent detection was performed using ECL reagent (Servicebio), and images were acquired with a LI-COR imaging system. Formaldehyde RNA immunoprecipitation (fRIP) fRIP was performed following modified protocols[ 24 , 25 ]. Briefly, mPFC tissue and subfractions were crosslinked with 0.1% formaldehyde, quenched with glycine, and lysed in native buffer with sonication (Covaris, 75W, 10% duty, 10 min). Lysates were incubated with specific antibodies and Protein G beads (Fisher Scientific) at 4°C. After washing, complexes were reverse-crosslinked with buffer containing proteinase K, DTT, and RNase inhibitor, followed by RNA purification (RNAClean XP, Beckman Coulter). Transmission electron microscope (TEM) Mice were perfused, and brains were fixed in 2.5% glutaraldehyde overnight at 4°C. Tissues were post-fixed in osmium tetroxide, dehydrated through graded ethanol and acetone, embedded in EPON812, and polymerized. ILPFC sections (400 nm) were stained with uranyl acetate and lead citrate and imaged using a Tecnai G2 F20 TEM (FEI, 80 kV). Golgi staining Golgi staining was performed using a commercial kit (Servicebio) following behavioral tests. Brains were immersed in impregnation solution for 15 days, sectioned at 40 µm, and stained. Dendritic morphology was analyzed with CaseViewer 2.4, NeuronJ, and Sholl Analysis plugins. Sequencing data analysis For both m¹A-RIP-seq and fRIP-seq, raw 150 bp paired-end reads were quality- controlled and adapter-trimmed using fastp (v0.23.2)[ 26 ], and rRNA-mapped reads were removed after alignment to rRNA reference sequences using HISAT2 (v2.2.1)[ 27 ]. Remaining reads were mapped to the mm39 mouse genome using STAR (v2.7.10b) [ 28 ]. BAM files were deduplicated (Sambamba, v0.8.2)[ 29 ] and filtered for high-quality alignments (SAMtools, v1.14)[ 30 ] with flags “-F 1804 -f 2 -q 20”. m¹A- and SRSF1-bound peaks were identified using exomePeak2 (v1.9.1)[ 31 ], which applies Wald tests on Poisson generalized linear models, with FDR correction via the Benjamini-Hochberg method. m¹A peak distribution along transcripts was visualized with Guitar (v2.10.0)[ 32 ]. Gene ontology (GO) analysis of m¹A-modified targets was performed with clusterProfiler (v3.18.1)[ 33 ], and enriched RNA motifs were identified using the MEME Suite (v5.5.2). For integrative analysis, overlap between m¹A-modified sites and SRSF1 peaks was computed using bedtools intersect (v2.30.0)[ 34 ]. Enrichment patterns were visualized via metagene plots using deepTools (v3.5.1)[ 35 ]. Correlations between m¹A and SRSF1 binding were assessed by binning the genome into 10 kb windows and quantifying read coverage with bedtools multicov followed by Pearson correlation analysis. Statistical analysis All statistical analyses were performed using GraphPad Prism 9. Most quantitative results are presented as box-and-whisker plots displaying individual values, median, and range; bar graphs are shown as mean ± SEM. For comparisons between two groups, data normality was first assessed using the Shapiro-Wilk test. If normality was not met, the nonparametric Mann-Whitney U test was used; otherwise, an unpaired two-tailed t-test was applied. Welch’s correction was used for t-tests when variances were unequal. For behavioral experiments, fear acquisition and extinction training data are presented as mean ± SEM, whereas retrieval test data are shown as box plots; all behavioral data were analyzed using two-way ANOVA followed by Tukey’s post hoc test for multiple group comparisons. Statistical significance was defined as follows: *p < 0.05; **p < 0.01; ***p < 0.001; ***p < 0.0001. Results m¹A accumulation in adult mPFC during extinction learning requires TRMT61A Given that fear extinction serves as the neurobiological foundation of exposure therapy, we asked whether m¹A plays an active role in extinction itself. Mice underwent fear acquisition training followed by either extinction (EXT) training or retention control (RC) the next day (Supplementary Fig. 1). Mass spectrometry revealed a significant increase in total m¹A levels in the mPFC after EXT training (Fig. 1 A). To investigate the potential regulator of m¹A accumulation during EXT, we examined known m¹A methyltransferase[ 13 ]. As TRMT61B is not expressed in mice, analysis focused on the TRMT6/TRMT61A complex[ 14 ]. Although TRMT6 and TRMT61A mRNA levels remained unchanged following EXT training (supplementary Fig. 2), their protein levels were significantly increased (Fig. 1 B, C). Given that TRMT61A is the catalytic subunit of the TRMT6/TRMT61A m¹A methyltransferase complex, we next designed a lentiviral shRNA construct targeting TRMT61A and stereotaxically delivered it into the ILPFC (Fig. 1 F). Knockdown efficiency was validated in both primary cortical neurons (PCNs) (Fig. 1 D). After fear acquisition (Fig. 1 G), mice were divided to ensure comparable freezing scores within each group. ILPFC shTRMT61A knockdown reduced total m¹A level (Fig. 1 E) and did not affect extinction training process (Fig. 1 H), but significantly impaired extinction memory retrieval (Fig. 1 I). Open field performance was comparable, indicating no effect on locomotion or anxiety (Supplementary Fig. 3). Together, these findings indicate that TRMT61A-mediated m¹A methylation is dynamically regulated by extinction learning and is required for the proper formation of extinction memory. Extinction reshapes the m¹A landscape toward synapses in the mPFC Since TRMT61A is required for extinction memory, we next mapped m¹A changes by m¹A RIP-seq in whole mPFC from RC and EXT mice. Motif analysis revealed conserved GA-rich consensus sequences in both groups, including a dominant AGGUAA motif (Fig. 2 A). Separate peak calling and overlap analysis identified 90 EXT-specific, 561 RC-specific, and 2,219 shared transcripts (Fig. 2 B). Metagene profiles were broadly similar between RC and EXT across the 5′UTR, CDS, and 3′UTR (Fig. 2 C). Gene ontology (GO) analysis of m¹A-modified genes in RC highlighted synapse-related processes, whereas EXT retained synaptic terms and additionally enriched RNA-regulatory pathways (Fig. 2 D), suggesting a redistribution of m¹A targets with extinction at the synapse. Motivated by this shift, we fractionated mPFC into synaptic and nuclear compartments using Percoll-sucrose gradients, Western blots verified clean separation (synapse: PSD95⁺/H3⁻; nucleus: PSD95⁻/H3⁺) (Supplementary Fig. 4). Despite no change in TRMT6 and TRMT61A mRNA between RC and EXT (Supplementary Fig. 5), protein levels were selectively elevated in synaptic fractions after extinction, with no nuclear effect (Fig. 2 E-G). Consistently, Ribosome Immunoprecipitation (Ribo IP)-qPCR showed greater ribosome association of TRMT6 and TRMT61A mRNAs at synapses in EXT, with no change in nuclear fractions (Supplementary Fig. 6), supporting synapse-specific local translation that may contribute to local m¹A accrual and synaptic remodeling. To further characterize compartment-specific regulation, we performed m¹A RIP-seq in synaptic and nuclear fractions from RC and EXT mPFC. All four conditions (RC-nucleus, EXT-nucleus, RC-synapse, EXT-synapse) showed GA-enriched potential m¹A motifs consistent with the whole-tissue analysis, while the EXT-synapse group exhibited a distinct GGAAGA motif (Fig. 2 H). In nuclear fractions, m¹A remained enriched at the 5’ UTR under both conditions, mirroring whole-tissue patterns; By contrast, extinction induced a redistribution of m¹A at synapses relative to nuclear and whole mPFC compartments, with reduced 5’ UTR enrichment and increased CDS/3’ UTR peaks (Fig. 2 J). GO terms overlapped across compartments (distal axon and myelin sheath), but nucleus-biased targets were preferentially associated with postsynaptic and dendritic spine-related processes, whereas synaptic m¹A-modified transcripts were selectively enriched for neurotransmitter transport, synaptic vesicle transport, and glutamatergic signaling (Fig. 2 K; Supplementary Fig. 7). A visual summary further highlighted increased synaptic 3′UTR-enriched peaks following extinction (Fig. 2 L). Given the regulatory importance of the 3′UTR in post-transcriptional control[ 36 – 38 ], we focused on synaptic transcripts with extinction-induced m¹A accumulation specifically in this region. Among 88 EXT-specific synaptic m¹A-modified transcripts (Fig. 2 I), Pten emerged as a prominent candidate. Extinction produced a narrow, high-intensity m¹A peak within the Pten 3′UTR at synapses, with no other peaks across the transcript and no signal in the nuclear compartment (Fig. 2 M; Supplementary Fig. 8). This site-specific modification was validated by m¹A RIP-qPCR (Supplementary Fig. 9). Together, these findings demonstrate compartment- and region-specific reshaping of m¹A during extinction learning, with a synaptic bias coupled to increased TRMT6/61A at synapses. Synaptic knockdown of Pten impairs fear extinction memory Protein analysis revealed that extinction training significantly increased PTEN expression in both total and synaptic mPFC fractions, whereas nuclear Pten levels remained unchanged (Fig. 3 A-C), implicating a synapse-specific regulatory mechanism. To directly assess the functional relevance of synaptic Pten , we employed a CRISPR-Cas-Inspired RNA Targeting System (CIRTS)-based RNA knockdown approach incorporating the Calm3 intron, which recruits RNA binding protein Staufen2 to enable selective targeting of synaptic transcripts[ 21 ]. Using a guide RNA specific to Pten mRNA, this system delivered the PIN RNase to selectively degrade Pten mRNA at synapse (Fig. 3 D). CIRTS system selectively reduced synaptic Pten mRNA in both the ILPFC and PCNs, with minimal impact on nuclear Pten levels (Fig. 3 E, F). Functionally, as outlined in the experimental workflow (Fig. 3 G), synaptic knockdown of Pten impaired extinction learning, reflected by altered extinction curves and a specific post hoc difference at the 9th CS presentation (Fig. 3 H). In contrast, baseline freezing in a novel context without extinction was unaffected (Fig. 3 I, PreCS), indicating that the effect was specific to extinction learning rather than baseline freezing. Notably, knockdown mice exhibited significantly increased freezing 24 hours after training (Fig. 3 I), indicating impaired retrieval and confirming a disruption in extinction memory consolidation. To exclude potential confounds, we next evaluated anxiety and locomotion. In the open field test, knockdown had no effect on center time, center entries, or total distance traveled (Supplementary Fig. 10A-C). In the elevated plus maze, open arm entries and time spent in open arms were similarly unaffected (Supplementary Fig. 10D, E). These findings support a specific role for synaptic Pten in extinction memory, rather than general changes in anxiety or locomotor activity levels. Together, these findings identify synaptic PTEN upregulation as a critical component of extinction memory formation, raising the possibility that m¹A may contribute to this process by modulating Pten at the post-transcriptional level. m¹A stabilizes synaptic Pten mRNA, promoting its protein accumulation and extinction memory Given that synaptic Pten is required for both the acquisition and retrieval of extinction memory, we next investigated the underlying mechanism. Specifically, we asked whether m¹A methylation at the 3’ UTR of Pten mRNA contributes to this regulation by promoting transcript stability and facilitating local protein synthesis during extinction. In PCNs, TRMT61A knockdown significantly reduced Pten mRNA stability under depolarizing (20 mM KCl) conditions (Fig. 4 A), suggesting that m¹A promotes Pten stability under activity-dependent conditions. Consistently, m¹A RIP-qPCR revealed a marked decrease in m¹A enrichment at the Pten 3’ UTR in TRMT61A-deficient neurons and ILPFC (Fig. 4 B), confirming that this modification is TRMT61A-dependent. These findings indicate that m¹A enhances Pten mRNA stability, which promote its protein accumulation during extinction retrieval. To assess the functional relevance of this m¹A event in vivo, we engineered a synapse-specific demethylation system by fusing ALKBH3 to the CIRTS-Calm3 system (Fig. 4 C). This construct selectively removed m¹A from the 3’ UTR of Pten mRNA in synaptic compartments of the ILPFC, with no detectable effect in nuclear fractions (Fig. 4 E). As outlined in the experimental workflow (Fig. 4 D), synaptic depletion of Pten m¹A did not impair extinction learning (Fig. 4 F), but disrupted extinction memory retrieval (Fig. 4 G). Baseline freezing behavior (Fig. 4 G, PreCS) and exploratory activity (Supplementary Fig. 11) remained unaffected, partially resembling the behavioral phenotype observed following synaptic Pten knockdown. Synaptic depletion of Pten m¹A impaired extinction retrieval, with anxiety-like behavior and locomotion unchanged. The convergent behavioral deficits after synaptic Pten knockdown and targeted m¹A demethylation thus support a synapse-localized m¹A to Pten mechanism in extinction memory. Extinction learning selectively recruits synaptic m¹A readers including SRSF1 RNA modifications can regulate gene expression via recruitment of RBPs. To identify RBPs recognizing m¹A-modified transcripts during extinction, we performed pull-downs with synthetic oligos containing a canonical m¹A motif from RIP-seq and prior reports[ 19 , 20 ]. Matched unmodified oligos served as controls, and bound proteins were analyzed by mass spectrometry (Fig. 5 A). We applied stringent criteria across biological replicates-consistent detection in EXT-synapse m¹A pull-downs, preferential enrichment over A-oligos in EXT but not RC conditions, and moderate log₂(m¹A/A) fold enrichment (1–8), identified six RBPs: SRSF1, MYEF2, ALYREF, FBL, DDX28, and RPL3 (Fig. 5 B). These proteins have been associated with diverse neuronal processes, including synaptic RNA metabolism[ 39 ], myelination[ 40 ], mRNA export[ 41 ], and activity-dependent translation[ 42 ]. Their selective enrichment in EXT-synapse fractions indicates extinction-induced recruitment of neuronally relevant m¹A readers (Fig. 5 C). Notably, unlike canonical readers YTHDF1-3 and YTHDC1, which were absent from synaptic fractions and not enriched in nuclear fractions (Fig. 5 D), extinction recruits a distinct set of synapse-localized m¹A readers. We next aimed to define the m¹A reader that binds Pten ’ s 3’ UTR and could couple methylation to function. Using RBPmap database, we screened the sequence surrounding the Pten synaptic m¹A peak and identified 93 candidate RBPs. Cross-referencing with the literature narrowed this to CPEB1, IGF2BP1, IGF2BP2, MSI1, RBM24, RBM38, and SRSF1protein, all previously reported to interact with Pten mRNA. SRSF1 uniquely overlapped both prediction and MS-based identification, implicating it as a likely m¹A-dependent interactor. We performed fRIP-qPCR and fRIP-seq to confirm these interactions (workflow in Fig. 5 E). fRIP-qPCR showed that extinction enhanced SRSF1 binding to the Pten 3’ UTR in synaptic, but not nuclear fractions of mPFC (Fig. 5 F). This interaction was reduced by CIRTS-ALKBH3 demethylation in PCNs (Fig. 5 G), confirming m¹A-dependence. To assess broader association, we performed SRSF1 fRIP-seq in whole mPFC and compared peaks with m¹A RIP-seq. The substantial overlap of SRSF1-bound RNAs with m¹A-modified transcripts (Fig. 5 H) further supports a broader role for SRSF1 in recognizing m¹A-modified RNAs. Together, these findings identify SRSF1 as an extinction-induced, synapse-enriched m¹A reader that binds Pten in an m¹A-dependent manner and may mediate broader post-transcriptional regulation during fear extinction. Synaptic m¹A depletion of Pten impairs dendritic and synaptic structure m¹A at the Pten 3’ UTR stabilizes the transcript and promote its protein accumulation during extinction, contributing to fear extinction memory, but whether it also drives synaptic structural remodeling remains unclear. In parallel, prior evidence implicating Pten in regulating synaptic architecture[ 43 – 46 ], prompting us to investigate whether synaptic depletion of m¹A at the 3’UTR of Pten leads to comparable morphological alterations. Using the CIRTS-ALKBH3 system to selectively demethylate the 3’ UTR of Pten mRNA at synapses, we assessed dendritic architecture in the ILPFC following extinction training. Golgi staining revealed a marked reduction in total dendritic length and branch complexity in mice with synaptic m¹A depletion compared to controls (Fig. 6 A-C). Dendritic spine analysis further revealed a significant reduction in spine density (Fig. 6 D-E), suggesting compromised structural plasticity. Given the observed dendritic deficits and the synaptic enrichment of vesicle-related transcripts identified in our m¹A RIP-seq data, we next examined presynaptic ultrastructure via transmission electron microscopy (TEM). Synaptic m¹A depletion of Pten reduced postsynaptic density (PSD) thickness, synaptic vesicle (SV) density, and increased cleft width, with no change in active zone (AZ) length (Fig. 6 H-L). Consistently, levels of synaptic proteins PSD95 and SYN1 were significantly reduced in ILPFC lysates from the m¹A-depleted group (Fig. 6 F-G). Together, these findings demonstrate that extinction-induced m 1 A modification of Pten mRNA at synapses supports not only behavioral plasticity but also the structural integrity of dendrites and synapses. This provides a mechanistic link between localized epitranscriptomic regulation and the anatomical remodeling required for extinction memory formation. Discussion m¹A is functionally implicated in diverse cellular processes, including tumor progression, stress responses, and oxidative adaptation[ 19 , 20 , 47 – 52 ]. However, its role in spatially defined RNA regulation within the adult brain, particularly at synapses relevant for learning and memory, remains uncharacterized. Although synaptic m⁶A accumulation has been linked to fear extinction[ 11 ], the potential involvement of m¹A in this form of plasticity is still unknown. We uncover a synapse-specific regulatory mechanism in which extinction training triggers TRMT6 and TRMT61A accumulation at synapses, leading to redistribution of m¹A marks and selective 3’ UTR methylation of Pten mRNA. This m¹A modification promotes Pten transcript stability and enhances protein expression through recruitment of SRSF1, identified here as a synaptic m¹A reader. These results define a mechanism in which TRMT61A installs m¹A to promote SRSF1 binding on Pten mRNA, thereby translating experience into molecular signals governing synaptic plasticity. Our findings suggest that m¹A is dynamically regulated by experience and may contribute to memory flexibility impairments observed in PTSD. In mice, extinction training elevated both m¹A abundance and TRMT6/61A expression in the mPFC, whereas knockdown of TRMT61A in the ILPFC impaired extinction memory, highlighting a functional role of m¹A in behavioral adaptation. At the transcriptomic level, extinction was accompanied by widespread remodeling of m¹A marks in the mPFC, particularly on genes related to synaptic organization and signaling. Together, these results suggest that m¹A promotes extinction by tuning synaptic gene expression networks. Motivated by this, we observed synapse-specific upregulation of TRMT6 and TRMT61A after extinction, which led us to investigate whether m¹A distribution similarly exhibits compartmental specificity. Subsequent subcellular m¹A RIP-seq revealed that synaptic m¹A preferentially enriched at coding sequences and 3’ UTRs, a compartment-specific pattern absents in nuclear or whole-tissue profiles. Together, these findings suggest that extinction induces synapse-specific m¹A remodeling to support local post-transcriptional regulation essential for fear extinction, prompting us to identify the relevant downstream effectors. Building on the role of 3’ UTRs in post-transcriptional regulation[ 36 – 38 ], we focused on synaptic transcripts with extinction-induced m¹A enrichment in this region. Among these, Pten mRNA was particularly notable, consistent with its known involvement in synaptic plasticity and cognitive function[ 53 , 54 ]. Using a CIRTS-PIN-Calm3 system, we performed synapse-specific knockdown of Pten in the ILPFC, which impaired extinction memory formation. This demonstrates the necessity of synaptic Pten for behavioral plasticity. Interestingly, our prior work in the basolateral amygdala (BLA) suggested that PTEN negatively regulates extinction[ 55 ], in apparent contrast to our observations in the ILPFC. This divergence likely reflects region- and circuit-level specificity in extinction control by PTEN. It may also arise from cell-type-specific effects: PTEN restricts dendritic growth in excitatory neurons, whereas its loss in interneurons accelerates maturation and limits plasticity via increased perineuronal nets[ 46 ]. In addition, manipulation scale matters: systemic inhibition of the PI3K-mTOR pathway impairs extinction[ 56 ], but such broad interventions differ fundamentally from our synapse-specific perturbation, which selectively disrupted extinction acquisition, highlighting the importance of spatial and cellular precision when modulating PTEN function. Finally, developmental stage further shapes outcomes: while Pten loss promotes synaptogenesis in juveniles [ 46 , 57 ], in our adult IL circuits Pten stabilizes synapses. We therefore propose that, in the mature ILPFC, PTEN supports extinction by maintaining synaptic integrity, thereby contributing to the regulation of extinction. To determine whether m¹A directly regulates synaptic Pten , we used the CIRTS-ALKBH3-Calm3 system to selectively demethylate its 3’ UTR. This site-specific removal of m¹A recapitulated the behavioral impairment observed with synaptic Pten knockdown. Since RNA modifications modulate transcript fate by affecting stability and translation[ 58 ], we further examined this mechanism. Demethylation reduced Pten mRNA stability in neurons, suggesting that m¹A enhances transcript longevity and supports protein expression, which is consistent with extinction-induced upregulation of synaptic PTEN protein. RNA modifications are known to function via recruitment of RNA-binding proteins (RBPs)[ 59 ]. In line with this, the synaptic enrichment of m¹A observed here may reflect selective engagement of RBPs during extinction learning. Using synthetic probes containing m¹A motifs, we performed RNA pull-down assays and identified six candidate RBPs enriched in the EXT-synapse condition. Among these, SRSF1 emerged as a top hit. Notably, previous studies have shown that SRSF1 binds the 3' UTR of Pten mRNA and enhances its stability and translation[ 60 ], consistent with our findings that m¹A promotes synaptic PTEN expression. We further demonstrated that in EXT process, this interaction is disrupted by m¹A demethylation, confirming that SRSF1 selectively engages m¹A-modified transcripts in a methylation-dependent manner. Importantly, whole-mPFC SRSF1 fRIP-seq revealed substantial overlap between SRSF1-bound regions and m¹A-modified transcripts in extinction-trained mice, indicating that SRSF1 functions not only as a noncanonical m¹A reader, but also as a core effector of synaptic RNA methylation in behaviorally relevant neural circuits. Together, these findings position SRSF1 as a molecular interpreter of synaptic m¹A marks, translating RNA methylation signals into transcript stabilization and local protein synthesis. This reveals a critical role for SRSF1 in bridging epitranscriptomic regulation with activity-dependent synaptic plasticity. Given prior evidence implicating PTEN in regulating synaptic architecture[ 43 – 46 ], our results demonstrate that extinction-induced, m¹A-dependent stabilization of Pten mRNA contributes to structural plasticity in the ILPFC. Selective demethylation of the m¹A site at the 3’ UTR of Pten led to pronounced impairments in dendritic morphology, including reduced arbor length, diminished branching complexity, and decreased spine density. Transmission electron microscopy further revealed attenuated PSD thickness, reduced synaptic vesicle density, and widened synaptic clefts, hallmarks of compromised synaptic ultrastructure, alongside decreased expression of core synaptic proteins such as SYN1 and PSD95. These data demonstrate that m¹A-mediated regulation of Pten is essential for maintaining synaptic integrity in extinction-relevant cortical circuits. By linking epitranscriptomic control to structural remodeling, this mechanism provides a molecular substrate through which experience modulates synaptic architecture to enable adaptive memory formation. In summary, our findings delineate a synapse-specific TRMT61A-dependent m¹A methylation at synapses promotes SRSF1 recruitment to Pten mRNA, links RNA methylation to local transcript stabilization, protein synthesis, and extinction memory formation. Through extinction-induced redistribution of m¹A and selective 3’ UTR methylation of Pten mRNA, this pathway enhances synaptic PTEN expression via SRSF1 recruitment, supporting structural and behavioral plasticity. These results identify synaptic m¹A remodeling as a mechanistic contributor to memory flexibility and suggest its dysregulation may underlie cognitive impairments in disorders such as PTSD. In line with this possibility, plasma m¹A levels were higher in healthy controls (HC) than in PTSD patients (Supplementary Fig. 12), paralleling the trend seen in extinction-trained versus retrieval control mice. By uncovering a spatially precise layer of post-transcriptional control, this work provides a framework for future studies of RNA-based mechanisms in experience-dependent neural adaptation. Declarations Conflict of Interest The authors declare no competing financial interests. Acknowledgements The authors are grateful to Prof. Timothy W. Bredy (University of Queensland) for kindly providing the CIRTS-PIN-Calm3 plasmid. This study was supported by the National Natural Science Foundation of China (Grant No. 82171517; 82271556; 82471534), the Major State Research Development Program of Hubei (Grant No. 2024BCA003) and Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province (No. 2023ZQNZD016). References Kessler RC, Aguilar-Gaxiola S, Alonso J, Benjet C, Bromet EJ, Cardoso G, et al. Trauma and PTSD in the WHO World Mental Health Surveys. Eur J Psychotraumatol. 2017;8(sup5):1353383. Koenen KC, Ratanatharathorn A, Ng L, McLaughlin KA, Bromet EJ, Stein DJ, et al. Posttraumatic stress disorder in the World Mental Health Surveys. Psychol Med. 2017;47(13):2260–74. Charlson FJ, Flaxman A, Ferrari AJ, Vos T, Steel Z, Whiteford HA. Post-traumatic stress disorder and major depression in conflict-affected populations: an epidemiological model and predictor analysis. Glob Ment Health (Camb). 2016;3:e4. Flores A, Fullana MA, Soriano-Mas C, Andero R. Lost in translation: how to upgrade fear memory research. Mol Psychiatry. 2018;23(11):2122–32. Ressler KJ, Berretta S, Bolshakov VY, Rosso IM, Meloni EG, Rauch SL, et al. Post-traumatic stress disorder: clinical and translational neuroscience from cells to circuits. Nat Rev Neurol. 2022;18(5):273–88. Peterson AL, Blount TH, Foa EB, Brown LA, McLean CP, Mintz J, et al. Massed vs Intensive Outpatient Prolonged Exposure for Combat-Related Posttraumatic Stress Disorder: A Randomized Clinical Trial. JAMA Netw Open. 2023;6(1):e2249422. Milad MR, Quirk GJ. Neurons in medial prefrontal cortex signal memory for fear extinction. Nature. 2002;420(6911):70–4. Sun C, Nold A, Fusco CM, Rangaraju V, Tchumatchenko T, Heilemann M, et al. The prevalence and specificity of local protein synthesis during neuronal synaptic plasticity. Sci Adv. 2021;7(38):eabj0790. van Oostrum M, Schuman EM. Understanding the molecular diversity of synapses. Nat Rev Neurosci. 2025;26(2):65–81. Nainar S, Marshall PR, Tyler CR, Spitale RC, Bredy TW. Evolving insights into RNA modifications and their functional diversity in the brain. Nat Neurosci. 2016;19(10):1292–8. Madugalle SU, Liau WS, Zhao Q, Li X, Gong H, Marshall PR, et al. Synapse-Enriched m(6)A-Modified Malat1 Interacts with the Novel m(6)A Reader, DPYSL2, and Is Required for Fear-Extinction Memory. J Neurosci. 2023;43(43):7084–100. Zhang C, Jia G. Reversible RNA Modification N(1)-methyladenosine (m(1)A) in mRNA and tRNA. Genomics Proteomics Bioinformatics. 2018;16(3):155–61. Li X, Xiong X, Zhang M, Wang K, Chen Y, Zhou J, et al. Base-Resolution Mapping Reveals Distinct m(1)A Methylome in Nuclear- and Mitochondrial-Encoded Transcripts. Mol Cell. 2017;68(5):993–1005 e9. Anderson J, Phan L, Hinnebusch AG. The Gcd10p/Gcd14p complex is the essential two-subunit tRNA(1-methyladenosine) methyltransferase of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A. 2000;97(10):5173–8. Chen M, Chen Y, Wang K, Deng X, Chen J. [Not Available]. Clin Transl Med. 2024;14(6):e1666. Li X, Xiong X, Wang K, Wang L, Shu X, Ma S, et al. Transcriptome-wide mapping reveals reversible and dynamic N(1)-methyladenosine methylome. Nat Chem Biol. 2016;12(5):311–6. Dominissini D, Nachtergaele S, Moshitch-Moshkovitz S, Peer E, Kol N, Ben-Haim MS, et al. The dynamic N(1)-methyladenosine methylome in eukaryotic messenger RNA. Nature. 2016;530(7591):441–6. Safra M, Sas-Chen A, Nir R, Winkler R, Nachshon A, Bar-Yaacov D, et al. The m1A landscape on cytosolic and mitochondrial mRNA at single-base resolution. Nature. 2017;551(7679):251–5. Zhang C, Yi X, Hou M, Li Q, Li X, Lu L, et al. The landscape of m(1)A modification and its posttranscriptional regulatory functions in primary neurons. Elife. 2023;12. Qi Z, Zhang C, Jian H, Hou M, Lou Y, Kang Y, et al. N(1)-Methyladenosine modification of mRNA regulates neuronal gene expression and oxygen glucose deprivation/reoxygenation induction. Cell Death Discov. 2023;9(1):159. Liau WS, Zhao Q, Bademosi A, Gormal RS, Gong H, Marshall PR, et al. Fear extinction is regulated by the activity of long noncoding RNAs at the synapse. Nat Commun. 2023;14(1):7616. Wei W, Zhao Q, Wang Z, Liau WS, Basic D, Ren H, et al. ADRAM is an experience-dependent long noncoding RNA that drives fear extinction through a direct interaction with the chaperone protein 14-3-3. Cell Rep. 2022;38(12):110546. Dunkley PR, Jarvie PE, Robinson PJ. A rapid Percoll gradient procedure for preparation of synaptosomes. Nat Protoc. 2008;3(11):1718–28. Selth LA, Gilbert C, Svejstrup JQ. RNA immunoprecipitation to determine RNA-protein associations in vivo. Cold Spring Harb Protoc. 2009;2009(6):pdb prot5234. D GH, Kelley DR, Tenen D, Bernstein B, Rinn JL. Widespread RNA binding by chromatin-associated proteins. Genome Biol. 2016;17:28. Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884-i90. Zhang Y, Park C, Bennett C, Thornton M, Kim D. Rapid and accurate alignment of nucleotide conversion sequencing reads with HISAT-3N. Genome Res. 2021;31(7):1290–5. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21. Tarasov A, Vilella AJ, Cuppen E, Nijman IJ, Prins P. Sambamba: fast processing of NGS alignment formats. Bioinformatics. 2015;31(12):2032–4. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2). Meng J, Lu Z, Liu H, Zhang L, Zhang S, Chen Y, et al. A protocol for RNA methylation differential analysis with MeRIP-Seq data and exomePeak R/Bioconductor package. Methods. 2014;69(3):274–81. Cui X, Wei Z, Zhang L, Liu H, Sun L, Zhang SW, et al. Guitar: An R/Bioconductor Package for Gene Annotation Guided Transcriptomic Analysis of RNA-Related Genomic Features. Biomed Res Int. 2016;2016:8367534. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3):100141. Quinlan AR. BEDTools: The Swiss-Army Tool for Genome Feature Analysis. Curr Protoc Bioinformatics. 2014;47:11 2 1–34. Ramirez F, Dundar F, Diehl S, Gruning BA, Manke T. deepTools: a flexible platform for exploring deep-sequencing data. Nucleic Acids Res. 2014;42(Web Server issue):W187-91. Bushkin GG, Pincus D, Morgan JT, Richardson K, Lewis C, Chan SH, et al. m(6)A modification of a 3' UTR site reduces RME1 mRNA levels to promote meiosis. Nat Commun. 2019;10(1):3414. Matoulkova E, Michalova E, Vojtesek B, Hrstka R. The role of the 3' untranslated region in post-transcriptional regulation of protein expression in mammalian cells. RNA Biol. 2012;9(5):563–76. Wang X, Zhao BS, Roundtree IA, Lu Z, Han D, Ma H, et al. N(6)-methyladenosine Modulates Messenger RNA Translation Efficiency. Cell. 2015;161(6):1388–99. Hulse RP, Drake RA, Bates DO, Donaldson LF. The control of alternative splicing by SRSF1 in myelinated afferents contributes to the development of neuropathic pain. Neurobiol Dis. 2016;96:186–200. Wang Y, Zhang Y, Zhang S, Kim B, Hull VL, Xu J, et al. PARP1-mediated PARylation activity is essential for oligodendroglial differentiation and CNS myelination. Cell Rep. 2021;37(1):109695. Shi M, Zhang H, Wu X, He Z, Wang L, Yin S, et al. ALYREF mainly binds to the 5' and the 3' regions of the mRNA in vivo. Nucleic Acids Res. 2017;45(16):9640–53. Giona F, Beretta S, Zippo A, Stefanoni A, Tomasoni Z, Vicidomini C, et al. Shank3 modulates Rpl3 expression and protein synthesis via mGlu5: implications for Phelan McDermid syndrome. Mol Psychiatry. 2025;30(8):3599–614. Knafo S, Sanchez-Puelles C, Palomer E, Delgado I, Draffin JE, Mingo J, et al. PTEN recruitment controls synaptic and cognitive function in Alzheimer's models. Nat Neurosci. 2016;19(3):443–53. Kwon CH, Luikart BW, Powell CM, Zhou J, Matheny SA, Zhang W, et al. Pten regulates neuronal arborization and social interaction in mice. Neuron. 2006;50(3):377–88. Tariq K, Cullen E, Getz SA, Conching AKS, Goyette AR, Prina ML, et al. Disruption of mTORC1 rescues neuronal overgrowth and synapse function dysregulated by Pten loss. Cell Rep. 2022;41(5):111574. D'Amore A, Sundberg M, Lin R, Lubbers ET, Winden KD, Yu L, et al. Phenotypic rescue via mTOR inhibition in neuron-specific Pten knockout mice reveals AKT and mTORC1-site specific changes. Mol Psychiatry. 2025;30(7):3077–89. Ueda Y, Ooshio I, Fusamae Y, Kitae K, Kawaguchi M, Jingushi K, et al. AlkB homolog 3-mediated tRNA demethylation promotes protein synthesis in cancer cells. Sci Rep. 2017;7:42271. Wang Y, Wang J, Li X, Xiong X, Wang J, Zhou Z, et al. N(1)-methyladenosine methylation in tRNA drives liver tumourigenesis by regulating cholesterol metabolism. Nat Commun. 2021;12(1):6314. Woo HH, Chambers SK. Human ALKBH3-induced m(1)A demethylation increases the CSF-1 mRNA stability in breast and ovarian cancer cells. Biochim Biophys Acta Gene Regul Mech. 2019;1862(1):35–46. Yamato I, Sho M, Shimada K, Hotta K, Ueda Y, Yasuda S, et al. PCA-1/ALKBH3 contributes to pancreatic cancer by supporting apoptotic resistance and angiogenesis. Cancer Res. 2012;72(18):4829–39. Wu Y, Chen Z, Xie G, Zhang H, Wang Z, Zhou J, et al. RNA m(1)A methylation regulates glycolysis of cancer cells through modulating ATP5D. Proc Natl Acad Sci U S A. 2022;119(28):e2119038119. Alriquet M, Calloni G, Martinez-Limon A, Delli Ponti R, Hanspach G, Hengesbach M, et al. The protective role of m1A during stress-induced granulation. J Mol Cell Biol. 2021;12(11):870–80. Knafo S, Esteban JA. PTEN: Local and Global Modulation of Neuronal Function in Health and Disease. Trends Neurosci. 2017;40(2):83–91. Chwin N, Kiryk A. The integral role of PTEN in brain function: from neurogenesis to synaptic plasticity and social behavior. Acta Neurobiol Exp (Wars). 2025;84(4):309–18. Murphy CP, Li X, Maurer V, Oberhauser M, Gstir R, Wearick-Silva LE, et al. MicroRNA-Mediated Rescue of Fear Extinction Memory by miR-144-3p in Extinction-Impaired Mice. Biol Psychiatry. 2017;81(12):979–89. Kritman M, Maroun M. Inhibition of the PI3 kinase cascade in corticolimbic circuit: temporal and differential effects on contextual fear and extinction. Int J Neuropsychopharmacol. 2013;16(4):825–33. Perandones C, Costanzo RV, Kowaljow V, Pivetta OH, Carminatti H, Radrizzani M. Correlation between synaptogenesis and the PTEN phosphatase expression in dendrites during postnatal brain development. Brain Res Mol Brain Res. 2004;128(1):8–19. Meyer KD, Jaffrey SR. The dynamic epitranscriptome: N6-methyladenosine and gene expression control. Nat Rev Mol Cell Biol. 2014;15(5):313–26. Roundtree IA, Evans ME, Pan T, He C. Dynamic RNA Modifications in Gene Expression Regulation. Cell. 2017;169(7):1187–200. Katsuyama T, Li H, Comte D, Tsokos GC, Moulton VR. Splicing factor SRSF1 controls T cell hyperactivity and systemic autoimmunity. J Clin Invest. 2019;129(12):5411–23. Additional Declarations The authors have declared there is NO conflict of interest to disclose Supplementary Files Supplementaryinformation.docx Supplementary Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7703002","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":523421075,"identity":"ffd8c13f-debd-4840-872d-0583b7f619ea","order_by":0,"name":"Xiang Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYBACPmYgkQDE/Aw8ID4zYS1sMC2SDURrgTEMDhCthZ2BTeJBzR27zTdyj0kwVFgnNrCfPUDIYcwGCceeJW+7kZcmwXAmPbGBJy+BkBbGBwlsh5PNbueYSTC2HU5skOAxIKSF4UDCv8PJxrNBWv4Rp4XxQWLbYTsDaZCWBqK0MDYbJPYdTpC4/8bYIuFYunEbTw5+Lfz8h49J/vh22J6/54zhjQ811rL97Gfwa2FgYGwAkYlgMoEBKaYIAXtiFY6CUTAKRsEIBAAA2zy86+1pWgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-6849-353X","institution":"Zhongnan Hospital of Wuhan University","correspondingAuthor":true,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Li","suffix":""},{"id":523421076,"identity":"1799f095-0fb9-41e0-b984-35ba67434776","order_by":1,"name":"Yi Zhang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Zhang","suffix":""},{"id":523421077,"identity":"ca4db113-c757-400a-a5fc-f651c31decb3","order_by":2,"name":"Jiazhi Jiang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jiazhi","middleName":"","lastName":"Jiang","suffix":""},{"id":523421078,"identity":"36cc6ebc-1b23-4426-b4e7-1e7a842b3d2f","order_by":3,"name":"Ziwei Pi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ziwei","middleName":"","lastName":"Pi","suffix":""},{"id":523421079,"identity":"57f46a4e-2b44-47e4-aa6d-bfb8adcd85e2","order_by":4,"name":"Tongyu Chen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tongyu","middleName":"","lastName":"Chen","suffix":""},{"id":523421080,"identity":"c19f6873-f194-42d0-aa3e-c177a5b27b1a","order_by":5,"name":"Zhipeng Xu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Zhipeng","middleName":"","lastName":"Xu","suffix":""},{"id":523421081,"identity":"19925c29-06ba-49c5-81d5-a6f4a1aa5f63","order_by":6,"name":"Yi Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Wang","suffix":""},{"id":523421082,"identity":"c553504f-745d-4117-ba13-c66bbe5e487f","order_by":7,"name":"Tian Lan","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tian","middleName":"","lastName":"Lan","suffix":""},{"id":523421085,"identity":"7ade9e50-9442-42e0-8be1-72b10d92ac97","order_by":8,"name":"Zhihan Sun","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Zhihan","middleName":"","lastName":"Sun","suffix":""},{"id":523421087,"identity":"a00e6812-c03b-455a-9cb9-0dcde7a36d8f","order_by":9,"name":"Junhui Liu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Junhui","middleName":"","lastName":"Liu","suffix":""},{"id":523421088,"identity":"04c7685f-5c81-476a-9966-65e4aa1e5485","order_by":10,"name":"Wei Wei","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Wei","suffix":""},{"id":523421089,"identity":"002cfca9-eb91-49af-b75a-229fa5257695","order_by":11,"name":"Xihai Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Xihai","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-09-24 11:10:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7703002/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7703002/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92698943,"identity":"1b9b3cd4-3983-46fe-b435-ca906aa3021b","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21465576,"visible":true,"origin":"","legend":"","description":"","filename":"fig3MPCIRTSPINcs3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/64dabb187afb9eb45b307f8a.tif"},{"id":92698941,"identity":"6f20f8be-e95b-468c-90e0-51428a5db1a3","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":171980,"visible":true,"origin":"","legend":"","description":"","filename":"manucriptMP.docx","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/66d21a3373c375abdcffd538.docx"},{"id":92699039,"identity":"397038f2-e259-4f8c-9261-1ce1365dbd54","added_by":"auto","created_at":"2025-10-03 07:48:22","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21443820,"visible":true,"origin":"","legend":"","description":"","filename":"fig4MPCIRTSALKBH33cs.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/3ed27138d6f670c6078fc0e9.tif"},{"id":92698951,"identity":"0d625548-0fd9-445a-ba1e-42eceab01852","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28125044,"visible":true,"origin":"","legend":"","description":"","filename":"fig5MPFRIP.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/a11c33f43165cf37f97bab77.tif"},{"id":92698949,"identity":"6729842c-481a-4710-acc7-4bbcd17615ea","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23815252,"visible":true,"origin":"","legend":"","description":"","filename":"fig6MPtem.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/91fe7144faf2505e24ebed35.tif"},{"id":92698948,"identity":"029df35c-70ec-43cc-bc17-da127d560b82","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31109316,"visible":true,"origin":"","legend":"","description":"","filename":"figrue1v4cs3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/550076317d40c96aaa45e1e3.tif"},{"id":92698950,"identity":"e67eb0fb-49da-4f62-8b38-036c3c949fbf","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29960584,"visible":true,"origin":"","legend":"","description":"","filename":"figure2MP.tif","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/6af161d8d2b9b38714d55b0d.tif"},{"id":92699038,"identity":"3cf3c855-8729-4672-9f05-b0146ff4f0a5","added_by":"auto","created_at":"2025-10-03 07:48:22","extension":"json","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13206,"visible":true,"origin":"","legend":"","description":"","filename":"2025MP002390.json","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/ece647d773a4059f7e7a9818.json"},{"id":92698952,"identity":"4f2502c7-fdea-4c79-aa85-726623ab4885","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"docx","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27503356,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/9f6264284c68b34dd88886cf.docx"},{"id":92698936,"identity":"5dbe37f5-15a4-4a37-9f8b-4e7121496238","added_by":"auto","created_at":"2025-10-03 07:40:21","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2146991,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTRMT61A-mediated m¹A accumulation is required for fear extinction memory.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Experimental workflow and m¹A quantification in mPFC of RC and EXT mice; EXT mice show increased m¹A levels (Two-tailed unpaired t test with Welch's correction, n=3, *\u003cem\u003ep\u003c/em\u003e = 0.0175). (B, C) TRMT61A and TRMT6 protein expression levels increase in mPFC after EXT training (n=4 per group, *\u003cem\u003ep\u003c/em\u003e= 0.018 and *\u003cem\u003ep\u003c/em\u003e = 0.012, respectively). (D) Knockdown efficiency of TRMT61A shRNA validated in primary cortical neurons (n=6 per group, ****\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.0001). (E) Dot blot shows reduced m¹A levels in ILPFC after TRMT61A knockdown (n=3, *\u003cem\u003ep\u003c/em\u003e = 0.0333). Statistical comparisons in panels B-E were performed using two-tailed unpaired Student’s t-tests. (F) Representative image of virus-infected ILPFC (GFP, green; DAPI, blue; scale bar: 1 mm). (G) Freezing percentage during fear conditioning was equivalent prior to virus injection. (H) TRMT61A knockdown does not impair within-session extinction learning (n=6-8, two-way ANOVA, F(1,12) = 0.8901, p = 0.3640; only the first 10 CS trials of the 30-trial session are shown). For panel G-H, error bars represent SEM. (I) TRMT61A knockdown impairs extinction memory but not general fear expression (two-way ANOVA, n = 6-8 as indicated in the figure, F(3,125) = 55.14, ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001; Tukey’s post hoc test: EXT control vs. EXT shTRMT61A, CS1 **p = 0.0019, CS2 ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001, CS3 p = 0.1677, avgCS **\u003cem\u003ep\u003c/em\u003e = 0.0029). CS, conditioned stimulus; preCS, a 2 min acclimation pretest period to minimize context generalization; CS1, CS2, CS3, response to the first, second and third of three tone CS exposures, at test; avgCS, average of 3 tone CS exposures. All control groups received a non-targeting shRNA.\u003c/p\u003e","description":"","filename":"figrue1v4cs3.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/82d9ca07c3f9a6a209bb974c.png"},{"id":92699037,"identity":"c0a4c03d-b9b8-48fe-8519-a6bd6bcc0613","added_by":"auto","created_at":"2025-10-03 07:48:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3613434,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSynapse-biased m¹A redistribution in the mPFC after extinction training, with elevated TRMT6/61A at synapses.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Predicted sequence motifs of m¹A peaks in mRNAs from RC and EXT groups. (B) Venn diagram showing m¹A-modified RNA-associated genes in mPFC of RC- and EXT-trained mice, with 561 RC-specific, 90 EXT-specific, and 2,219 shared genes. (C) Metagene plot of m¹A deposition across transcripts in the mPFC of RC and EXT mice. (D) Gene ontology analysis of the top five gene clusters associated with m¹A deposition differences between RC and EXT groups. (E–G) Protein levels of TRMT6 and TRMT61A in subcellular fractions of the mPFC. (E) Representative immunoblots for synaptic (GAPDH/ACTB loading controls) and nuclear fractions (H3 loading control), \u003cem\u003en\u003c/em\u003e = 4. (F) Synaptic fraction quantification shows increased TRMT6 (left; *\u003cem\u003ep \u003c/em\u003e= 0.0206) and TRMT61A (right; ***\u003cem\u003ep \u003c/em\u003e= 0.0001) after EXT training. (G) Nuclear fraction quantification shows no significant change in TRMT6 ( \u003cem\u003ep\u003c/em\u003e = 0.7909) or TRMT61A (\u003cem\u003ep\u003c/em\u003e = 0.5528). For panel F-G, two-tailed unpaired Student’s t-tests. (H) Predicted sequence motifs of m¹A peaks in nucleus (top) and synapse (bottom) from RC and EXT groups. (I) Venn diagrams of m¹A-modified RNA-associated genes in nucleus (top; 219 RC-specific, 340 EXT-specific, 599 shared) and synapse (bottom; 92 RC-specific, 88 EXT-specific, 73 shared). (J) Metagene plots of m¹A deposition across transcripts in synaptic and nuclear compartments of RC and EXT mice. (K) Gene ontology analysis of synaptic mRNAs differentially methylated between RC and EXT groups. (L) Distribution of m¹A peaks across transcript regions in synaptic mPFC of RC and EXT mice. Pie charts show proportions at 5’ UTR, CDS, start/stop codons, and 3’ UTR (RC 3’ UTR: 12.8%; EXT 3’ UTR: 17.6%); bar graphs depict feature enrichment ratios. Representative genes with synaptic 3’ UTR peaks in EXT are listed at right. (M) \u003cem\u003ePten\u003c/em\u003e transcript tracks showing differential methylation patterns at the synapse under RC versus EXT; yellow highlight indicates the differential position in the 3’ UTR.\u003c/p\u003e","description":"","filename":"figure2MP.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/9585bfadf9f61aef39594237.png"},{"id":92698937,"identity":"52381e7f-e4ec-4f1e-badd-34212719b1bf","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1559749,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTargeted synaptic \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePten\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e knockdown impairs fear extinction memory.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A–C) PTEN protein expression is increased in the mPFC (A, \u003cem\u003en\u003c/em\u003e=4, **\u003cem\u003ep\u003c/em\u003e = 0.0088) and synaptic fraction (B, \u003cem\u003en\u003c/em\u003e = 3 independent pooled samples, each pool from 4 mice, total 12 mice, *\u003cem\u003ep\u003c/em\u003e = 0.0307) following extinction training, with no significant change in the nuclear fraction (C, \u003cem\u003en\u003c/em\u003e=3, representing 12 biological replicates, \u003cem\u003ep\u003c/em\u003e= 0.8689). (D) Schematic of the CIRTS-PIN lentiviral construct used for synaptic-specific \u003cem\u003ePten\u003c/em\u003e knockdown. (E) In primary cortical neurons, CIRTS-PIN knockdown reduces Pten expression at the synapse (right, **\u003cem\u003ep\u003c/em\u003e = 0.0021), with no effect in the nucleus (left, \u003cem\u003ep\u003c/em\u003e = 0.4891). (F) In the ILPFC of mice, CIRTS-PIN-mediated knockdown reduces synaptic \u003cem\u003ePten\u003c/em\u003e (right, representing 12 biological replicates, *\u003cem\u003ep\u003c/em\u003e = 0.0464), but not nuclear levels (left, representing 12 biological replicates, \u003cem\u003ep\u003c/em\u003e = 0.8445). For panels D-F, \u003cem\u003en\u003c/em\u003e= 3. For panels A-F, two-tailed unpaired Student’s t-tests were used, error bars represent SEM. (G) Schematic of behavioral protocol used to assess the effect of \u003cem\u003ePten\u003c/em\u003eknockdown in ILPFC synapses on fear extinction memory. CTX: context; CS: conditioned stimulus; US: unconditioned stimulus. (H) Synaptic \u003cem\u003ePten\u003c/em\u003eknockdown impairs within-session extinction learning during EXT training (\u003cem\u003en\u003c/em\u003e=11-12, two-way ANOVA, F(1,21) = 12.71, column factor **\u003cem\u003ep\u003c/em\u003e = 0.0018, CS9 *\u003cem\u003ep\u003c/em\u003e= 0.0494, error bars represent SEM). (I) Synaptic \u003cem\u003epten\u003c/em\u003e knockdown has no effect on fear expression in RC-trained mice, but significantly impairs extinction memory in EXT-trained mice (n = 6–12 as indicated in the figure, two-way ANOVA, F(3,155) = 19.86, column factor ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001; Tukey’s post hoc: EXT Control vs. EXT gRNA,CS1 \u003cem\u003ep\u003c/em\u003e = 0.6806, CS2 **\u003cem\u003ep\u003c/em\u003e= 0.0078, CS3 **\u003cem\u003ep\u003c/em\u003e = 0.0041, avgCS *\u003cem\u003ep\u003c/em\u003e = 0.0495). All control groups received the same CIRTS-PIN vector carrying a non-targeting gRNA.\u003c/p\u003e","description":"","filename":"fig3MPCIRTSPINcs3.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/edda97de7aac8edd0a9af301.png"},{"id":92698939,"identity":"936ef338-f19d-4ac1-b615-af31a08e5365","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1464287,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTargeted m¹A demethylation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePten\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e at the synapse impairs fear extinction memory.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) \u003cem\u003ePten\u003c/em\u003e mRNA decay measured by actinomycin-D pulse-chase in primary cortical neurons (PCNs) stimulated with 20 mM KCl; TRMT61A knockdown accelerates degradation (\u003cem\u003en\u003c/em\u003e=3, two-way ANOVA, F(1,4) = 32.98, column factor **\u003cem\u003ep\u003c/em\u003e = 0.0046). (B) m¹A-RIP-qPCR shows reduced \u003cem\u003ePten\u003c/em\u003e m¹A modification following TRMT61A knockdown in PCNs (left, \u003cem\u003en\u003c/em\u003e=3, *\u003cem\u003ep\u003c/em\u003e = 0.0004) and ILPFC (right, \u003cem\u003en\u003c/em\u003e=3, *\u003cem\u003ep\u003c/em\u003e = 0.0005). two-tailed unpaired Student’s t-tests. Control groups received the same FG12 vector carrying a non-targeting shRNA. For panel A-B, Error bars represents SEM. (C) Schematic of CIRTS-ALKBH3 construct used for targeted m¹A demethylation of \u003cem\u003ePten\u003c/em\u003e. (D) Behavioral protocol for testing the effect of \u003cem\u003ePten\u003c/em\u003e m¹A demethylation in ILPFC synapses on fear extinction. CTX: context; CS: conditioned stimulus; US: unconditioned stimulus. (E) m¹A-RIP-qPCR of \u003cem\u003ePten\u003c/em\u003e in ILPFC synaptic and nuclear fractions shows reduced m¹A enrichment at the synapse (left, \u003cem\u003en\u003c/em\u003e=6, *\u003cem\u003ep\u003c/em\u003e = 0.0158) and no change in the nucleus (right, \u003cem\u003en\u003c/em\u003e=6, \u003cem\u003ep\u003c/em\u003e = 0.7058). two-tailed unpaired Student’s t-tests. (F) Targeted m¹A demethylation of \u003cem\u003ePten\u003c/em\u003e does not impair within-session extinction learning (\u003cem\u003en\u003c/em\u003e=7-11, two-way ANOVA, F(1,21) = 1.023, column factor \u003cem\u003ep\u003c/em\u003e = 0.3232). Error bars represent SEM. (G) While \u003cem\u003ePten\u003c/em\u003e demethylation does not affect fear memory expression in RC-trained mice, it significantly impairs extinction memory in EXT-trained mice (\u003cem\u003en\u003c/em\u003e=7-11 as indicated in the figure; two-way ANOVA, F(3,165) = 53.5, column factor ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001; Tukey’s post hoc test: EXT control vs. EXT gRNA, CS1 \u003cem\u003ep\u003c/em\u003e = 0.2204, CS2 \u003cem\u003ep\u003c/em\u003e = 0.0659, CS3 **\u003cem\u003ep\u003c/em\u003e= 0.0062, avgCS \u003cem\u003ep\u003c/em\u003e=0.0526). All control groups in panels E–G received the same CIRTS-ALKBH3 vector carrying a non-targeting gRNA.\u003c/p\u003e","description":"","filename":"fig4MPCIRTSALKBH33cs.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/cd4bd094e4dcc70a37de03d5.png"},{"id":92698942,"identity":"a76c45eb-268f-4a77-a575-4295749d8272","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":7365400,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRNA pull-down identifies synapse-enriched m¹A readers during extinction learning, with SRSF1 selectively binding \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePten\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e in an m¹A-dependent manner.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Schematic of RNA pull-down workflow for synaptic and nuclear fractions of the mPFC. (B) Enrichment of selected RNA-binding proteins (RBPs) on m¹A-modified oligos compared to unmodified controls, shown as log₂ fold change [(m¹A+1)/(A+1)] in RC and EXT synaptic samples (\u003cem\u003en\u003c/em\u003e=2, error bars represent SD). (C, D) Scatter plots of normalized protein abundance from m¹A/A pull-downs in synaptic (C) and nuclear (D) lysates. SRSF1 shows extinction-specific enrichment selectively in the synapse. (E) Experimental workflow for assessing SRSF1 binding to RNAs following fear conditioning (RC) or extinction (EXT). (F) fRIP-qPCR shows SRSF1 binding to m¹A peaks in the \u003cem\u003ePten\u003c/em\u003e 3’ UTR in mPFC nucleus (\u003cem\u003en\u003c/em\u003e=6, Mann–Whitney test, \u003cem\u003ep\u003c/em\u003e = 0.6667, ns) and synapse ( \u003cem\u003en\u003c/em\u003e=6, *\u003cem\u003ep\u003c/em\u003e = 0.019). two-tailed unpaired Student’s t-tests. (G) fRIP-qPCR in primary cortical neurons infected with CIRTS-ALKBH3 shows SRSF1 binding to the same \u003cem\u003ePten\u003c/em\u003e 3’ UTR site in synapse (\u003cem\u003en\u003c/em\u003e=3, *\u003cem\u003ep\u003c/em\u003e = 0.0437) but not in the nucleus (\u003cem\u003en\u003c/em\u003e=3, \u003cem\u003ep\u003c/em\u003e = 0.5577), Error bars represent SEM. Control groups received the same CIRTS-ALKBH3 vector carrying a non-targeting gRNA. For panel F-G, two-tailed unpaired Student’s t-tests were used. (H) SRSF1 fRIP-seq signals (top: metaplots; bottom: heatmaps) centered on m¹A peaks (±2 kb) in RC and EXT mPFC. Each row represents an m¹A peak; each column represents an individual biological replicate.\u003c/p\u003e","description":"","filename":"fig5MPFRIP.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/47b99c4ccf64a6b92b2cfd8f.png"},{"id":92698946,"identity":"1330340c-89a4-41e9-b3f7-23b6e875d155","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":7465747,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSynapse-specific knockdown of m¹A modification on the \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePten\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e 3’ UTR under extinction conditions alters ILPFC synapse structure.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative Golgi-stained neurons in ILPFC after injection of CIRTS-ALKBH3 Control or gRNA viruses. Scale bar: 50 µm. (B) Total dendritic length is reduced in the CIRTS-ALKBH3 gRNA group (\u003cem\u003en\u003c/em\u003e=10 neurons from 3 mice per group, two-tailed unpaired Student’s t-tests, *\u003cem\u003ep\u003c/em\u003e = 0.0121). (C) Sholl analysis shows reduced dendritic intersections in gRNA-injected mice (\u003cem\u003en\u003c/em\u003e=10 neurons from 3 mice per group, two-way ANOVA with Sidak’s multiple comparisons, F(1,360) = 49.69, ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001; significant pairwise differences observed at 35–50 µm, 35µm *\u003cem\u003ep\u003c/em\u003e = 0.0235, 40µm **\u003cem\u003ep\u003c/em\u003e = 0.0024, 45µm **\u003cem\u003ep\u003c/em\u003e= 0.0093, 50µm *\u003cem\u003ep\u003c/em\u003e = 0.0174；Error bars represent SEM). (D) Representative high-magnification Golgi-stained dendrites. Scale bar: 20 µm. (E) Dendritic spine density is reduced in the gRNA group (\u003cem\u003en\u003c/em\u003e=20 dendrites from 3 mice per group, two-tailed unpaired Student’s t-tests, ***\u003cem\u003ep\u003c/em\u003e = 0.0003). (F, G) Western blot showing decreased SYN-1 (*\u003cem\u003ep\u003c/em\u003e = 0.033) and PSD-95 (*\u003cem\u003ep\u003c/em\u003e = 0.0493) protein levels in ILPFC after CIRTS-ALKBH3 gRNA injection (\u003cem\u003en\u003c/em\u003e=3 per group)，two-tailed unpaired Student’s t-tests, Error bars represent SEM. (H) Representative TEM images of ILPFC synapses after CIRTS-ALKBH3 Control or gRNA injection. Scale bars: 2 µm (left), 500 nm (right). (I–L) Quantification of ultrastructural features shows reduced PSD thickness (\u003cem\u003en\u003c/em\u003e=10 synapses from 3 mice per group, ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001), no change in active zone (AZ) length (\u003cem\u003ep\u003c/em\u003e = 0.0559), reduced synaptic vesicle (SV) density (***\u003cem\u003ep \u003c/em\u003e= 0.0005), and increased synaptic cleft length (****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001) in the gRNA group, two-tailed unpaired Student’s t-tests. All control groups received the same CIRTS-ALKBH3 vector carrying a non-targeting gRNA\u003cem\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e","description":"","filename":"fig6MPtem.png","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/ebb1a9d6c96e7a785ab0f1ec.png"},{"id":92699678,"identity":"ecc8f7e9-341d-4b3b-97f2-09f674de72a4","added_by":"auto","created_at":"2025-10-03 07:56:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":23853013,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/1b66c9e9-48eb-494a-9810-b321d471396e.pdf"},{"id":92698944,"identity":"066e94a3-99a0-4920-b993-bffa52094772","added_by":"auto","created_at":"2025-10-03 07:40:22","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27503356,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7703002/v1/358dee7af330b6a3ffee7d80.docx"}],"financialInterests":"The authors have declared there is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Synaptic m1A remodeling in the mPFC promotes fear extinction via Pten regulation","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePost-traumatic stress disorder (PTSD) is a trauma-related psychiatric condition characterized by the persistence of intrusive fear memories and impaired regulation of threat responses, leading to significant psychological and social dysfunction (American Psychiatric Association, DSM-5). While approximately 70% of individuals experience traumatic events during their lifetime[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], only a subset (~\u0026thinsp;5.6%) develop PTSD[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], with lifetime prevalence estimates reaching up to 12.9% in certain populations[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], suggesting that individual differences in fear recovery mechanisms shape resilience or vulnerability. Among these, fear extinction, a form of inhibitory learning that enables the suppression of threat responses when danger is no longer present, has emerged as a core adaptive process disrupted in PTSD[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Accordingly, extinction-based psychotherapies, such as exposure therapy, represent the most effective interventions for PTSD[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. However, therapeutic efficacy remains limited: remission is achieved in only approximately 50% of individuals with combat-related PTSD, and relapse rates remain high[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. These limitations underscore a critical need to elucidate the molecular mechanisms underlying extinction learning and their dysfunction in PTSD.\u003c/p\u003e\u003cp\u003eAt the neural systems level, the medial prefrontal cortex (mPFC) plays a critical role in fear extinction by consolidating and retrieving fear extinction memory by inhibiting conditioned fear responses[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. At the cellular level, extinction learning is supported by synaptic plasticity, the activity-dependent remodeling of synapses that enables durable behavioral adaptation[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Given the spatial segregation between synapses and the nucleus, local RNA regulation is crucial for stimulus-evoked protein synthesis within synaptic compartments[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Despite this, the molecular mechanisms governing RNA stability and translation at synapses during extinction learning remain incompletely understood.\u003c/p\u003e\u003cp\u003eEpitranscriptomic regulation has recently emerged as an important mechanism for activity- and context-dependent gene expression in the brain[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. While N⁶-methyladenosine (m⁶A) has been intensively characterized in the brain, recent work showed that synapse-enriched m⁶A-modified Malat1 interacts with a novel m⁶A reader at synapses[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], whereas the roles of other RNA modifications remain poorly understood. Among these modifications, N\u0026sup1;-methyladenosine (m\u0026sup1;A), a reversible modification installed by the TRMT6-TRMT61A methyltransferase complex and removed by ALKBH3, has been identified across diverse RNA species[\u003cspan additionalcitationids=\"CR13 CR14 CR15\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. m\u0026sup1;A deposition at the 5\u0026rsquo; untranslated region (UTR) enhances cap-independent translation initiation, whereas its presence in coding sequences(CDS) may impede elongation due to its positive charge[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Notably, neuronal contexts exhibit a redistribution of m\u0026sup1;A toward the 3\u0026rsquo; UTR under stress conditions[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], suggesting a cell-type- and activity-specific reprogramming of m\u003csup\u003e1\u003c/sup\u003eA topology, though its functional role in synaptic compartments and behavior remains undefined.\u003c/p\u003e\u003cp\u003eHere we examine the role and distribution of m\u0026sup1;A in extinction learning. We found that m\u0026sup1;A was dynamically regulated in extinction-trained mice, while knockdown of TRMT61A abolished this regulation and impaired extinction memory. Subcellular profiling revealed synapse-specific remodeling of m\u0026sup1;A, nuclear m\u0026sup1;A retained 5\u0026rsquo; UTR enrichment regardless of condition, mirroring whole-tissue patterns; extinction triggered a synapse-specific shift from 5\u0026rsquo; UTR toward CDS and 3\u0026rsquo; UTR peaks. Among the modified transcripts, \u003cem\u003ePten\u003c/em\u003e emerged as a key synaptic target, regulated through TRMT61A-mediated m\u0026sup1;A and its recognition by SRSF1. Perturbation of this pathway disrupted dendritic architecture and impaired extinction memory, underscoring its behavioral significance. Together, our study uncovers an m\u0026sup1;A-centered regulatory pathway that links synaptic plasticity to extinction memory and provides new insight into the molecular basis of extinction memory formation.\u003c/p\u003e"},{"header":"Method","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePlasma collection and m\u0026sup1;A quantification\u003c/h2\u003e\u003cp\u003ePeripheral blood samples were obtained from PTSD patients and healthy controls, matched for age and sex, with informed consent and IRB approval. (Zhongnan Hospital, Wuhan University, Kelun-2022174). Plasma was separated by centrifugation (1000g, 10min, 4\u0026deg;C) and stored at -80\u0026deg;C until analysis. m\u0026sup1;A levels were quantified using a commercial ELISA kit (Cell Biolabs, MET-5099) and performed according to the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eAnimals\u003c/h3\u003e\n\u003cp\u003eMale C57BL/6J mice (9\u0026ndash;12 weeks) were housed under a 12-hour light/dark cycle with food and water ad libitum. All animal procedures were approved by the Animal Ethics Committee of Wuhan University (Approval No. ZN2023196).\u003c/p\u003e\n\u003ch3\u003eCell culture\u003c/h3\u003e\n\u003cp\u003ePrimary cortical neurons were isolated from E16 mouse embryos and cultured in Neurobasal medium (Gibco) supplemented with 1% GlutaMAX (Gibco), 2% B-27 (Gibco), and 1% penicillin/streptomycin (Servicebio, China). N2a and HEK293T cells were maintained in high-glucose DMEM ((Servicebio, China) with 10% FBS and 1% penicillin/streptomycin All kinds of cells were cultured at 37\u0026deg;C with 5% CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e\n\u003ch3\u003eCloning and lentiviral production\u003c/h3\u003e\n\u003cp\u003eCIRTS-PIN-Calm3 plasmid was obtained from the Timothy W. Bredy laboratory (The University of Queensland), with the cloning process detailed in previous publication[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. ALKBH3 coding sequence was cloned into the CIRTS backbone to generate CIRTS-ALKBH3. shRNA constructs targeting TRMT61A or scrambled control (Table\u0026nbsp;1) were cloned into FG12H1 vector. Lentiviruses were produced in HEK293T cells using PEI-mediated transfection and ultracentrifugation. Viral titers\u0026thinsp;\u0026gt;\u0026thinsp;1 \u0026times; 10^8 IU/mL were used.\u003c/p\u003e\n\u003ch3\u003eStereotactic viral injection\u003c/h3\u003e\n\u003cp\u003eMice were anesthetized with 1% pentobarbital (50 mg/kg, i.p.) and secured in a stereotaxic frame (RWD Life Science, Shenzhen, China). A heating-pad maintained body temperature during the procedure. Lentivirus (500 \u0026micro;L per side, 75 nL/min) was bilaterally injected into the ILPFC (2.00 mm anterior to bregma, 0.20 mm lateral to midline, and 2.75 mm ventral to dura) using a glass pipette and microinjection pump (RWD Life Science). The pipette remained in place for 10 minutes to prevent viral spread before the brain wound was sutured. Mice were returned to home cages after awakening and allowed at least one week for recovery and stable viral expression.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eBehavioral assays (fear conditioning, extinction and open field)\u003c/h2\u003e\u003cp\u003eTwo contexts (A and B) were used for fear conditioning and extinction behavioral testing. Conditioning chambers had transparent and stainless-steel walls with metal grid floors. In context B, the floor was covered with white plastic and lit by LED lights to reduce context generalization. Olfactory cues were also used to distinguish the contexts, with lemon scent applied in context A and vinegar in context B. Cameras recorded freezing behavior using Freezeframe software. Fear conditioning (FC) in context A included a 120 s pre-conditioning period, followed by three 120 s tones (80 dB, 16 kHz) paired with a 1 s, 0.6 mA foot shock. Mice were grouped based on freezing levels during the final tone and randomly assigned to different groups, ensuring comparable baseline freezing across groups. Extinction (EXT) in context B involved 30 non-reinforced 120 s tone presentations, while Retention Control (RC) mice were exposed to context B without tones. After 24 h, freezing was tested in context B during three 120 s tones, and memory was calculated as the percentage of freezing time.\u003c/p\u003e\u003cp\u003eOpen-field testing (OFT) was performed as described[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], with a shortened session (5 min).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTissue collection\u003c/h3\u003e\n\u003cp\u003eFor tissue collection, fear acquisition was performed as in the behavioral tests. After 24 hours of FC, mice were exposed to 60 cycles of tones (EXT group) or silence (RC group). Tissues were collected immediately afterward.\u003c/p\u003e\n\u003ch3\u003eSynaptosome preparation\u003c/h3\u003e\n\u003cp\u003emPFC tissue was homogenized and synaptosomes were isolated using Percoll gradient centrifugation as previously described[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Synaptosome purity was verified by Western blot.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eRNA isolation and RT-qPCR\u003c/h2\u003e\u003cp\u003eRNA was extracted using RNAiso Plus (Takara) and reverse-transcribed using PrimeScript Reverse Transcription Kit (Vazyme, China). Quantitative PCR was performed on a RotorGeneQ (Qiagen) cycler with SYBR Green Master Mix (Vazyme, China) and primers for target genes and PGK1 as a control (Supplementary Table). Transcript levels were normalized to PGK1 mRNA via the ΔΔCT method. Each PCR reaction was performed in duplicate for each sample and repeated at least twice. The primers used for RT-qPCR are listed in Table\u0026nbsp;1.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eLow input m1A-RIP seq\u003c/h2\u003e\u003cp\u003eThe experiment was performed as previously described[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], with minor modifications. Briefly, Protein G magnetic beads (Thermo Fisher) were incubated with m\u0026sup1;A-specific antibody (MBL, China) in m\u0026sup1;A binding buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 2 mM EDTA, 1 mM DTT, RNase inhibitor) for 1 h at room temperature. Total RNA was fragmented with 10 mM ZnCl₂ in 10 mM Tris-HCl at 94\u0026deg;C for 1 min. immediately quenched on ice with 0.5 M EDTA, and incubated with antibody-coupled beads overnight at 4\u0026deg;C. Beads were sequentially washed with m\u0026sup1;A buffer, low-salt, high-salt, and TET buffers. m\u0026sup1;A-enriched RNA was eluted at 42\u0026deg;C using DTT-containing elution buffer and purified with RNA Clean \u0026amp; Concentrator Kit (Zymo). Libraries were constructed with SMARTer Stranded Pico Input RNA Kit v2 (Takara) and sequenced on an Illumina platform (PE150).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eRNA pull-down and mass spectrometry\u003c/h2\u003e\u003cp\u003eBiotinylated RNA probes containing m\u0026sup1;A or unmodified adenosine were synthesized (5\u0026rsquo; Biotin-CCUGGGGG/m1A/AGACCCAGC-3\u0026rsquo; and 5\u0026rsquo; Biotin-CCUGGGGGAAGACCCAGC-3\u0026rsquo;) (TSINGKE, China). Nucleus-enriched and synaptosomal lysates from ILPFC of RC or EXT-trained mice were prepared in lysis buffer (10 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 10 mM Tris-HCl, pH 7.5, 0.5 mM DTT) with protease inhibitors (Roche). After preclearing with streptavidin beads (Thermo Fisher, #88817), lysates were incubated with RNA probes (2 \u0026micro;g) and captured on streptavidin beads pre-blocked with BSA and tRNA. Bound proteins were resolved by SDS-PAGE (10% Bis-Tris, Invitrogen), silver-stained (Solarbio), digested with trypsin, and identified by LC-MS/MS (Weizmann Institute MS Core). Two independent replicates were performed.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eProtein extraction and western blotting\u003c/h2\u003e\u003cp\u003eTissues or subcellular fractions were lysed in RIPA buffer (Servicebio) containing protease inhibitors. Protein concentration was measured by BCA assay (Beyotime), and lysates were denatured at 95\u0026deg;C for 10 min. Proteins were separated by SDS-PAGE and transferred to PVDF membranes (Millipore). After blocking, membranes were probed with primary antibodies against TRMT61A, TRMT6, PSD95, SYN1, PTEN, GAPDH, ACTB, and H3, followed by HRP-conjugated secondary antibodies (Table\u0026nbsp;1). Chemiluminescent detection was performed using ECL reagent (Servicebio), and images were acquired with a LI-COR imaging system.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eFormaldehyde RNA immunoprecipitation (fRIP)\u003c/h2\u003e\u003cp\u003efRIP was performed following modified protocols[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Briefly, mPFC tissue and subfractions were crosslinked with 0.1% formaldehyde, quenched with glycine, and lysed in native buffer with sonication (Covaris, 75W, 10% duty, 10 min). Lysates were incubated with specific antibodies and Protein G beads (Fisher Scientific) at 4\u0026deg;C. After washing, complexes were reverse-crosslinked with buffer containing proteinase K, DTT, and RNase inhibitor, followed by RNA purification (RNAClean XP, Beckman Coulter).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eTransmission electron microscope (TEM)\u003c/h2\u003e\u003cp\u003eMice were perfused, and brains were fixed in 2.5% glutaraldehyde overnight at 4\u0026deg;C. Tissues were post-fixed in osmium tetroxide, dehydrated through graded ethanol and acetone, embedded in EPON812, and polymerized. ILPFC sections (400 nm) were stained with uranyl acetate and lead citrate and imaged using a Tecnai G2 F20 TEM (FEI, 80 kV).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eGolgi staining\u003c/h2\u003e\u003cp\u003eGolgi staining was performed using a commercial kit (Servicebio) following behavioral tests. Brains were immersed in impregnation solution for 15 days, sectioned at 40 \u0026micro;m, and stained. Dendritic morphology was analyzed with CaseViewer 2.4, NeuronJ, and Sholl Analysis plugins.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eSequencing data analysis\u003c/h2\u003e\u003cp\u003eFor both m\u0026sup1;A-RIP-seq and fRIP-seq, raw 150 bp paired-end reads were quality- controlled and adapter-trimmed using fastp (v0.23.2)[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], and rRNA-mapped reads were removed after alignment to rRNA reference sequences using HISAT2 (v2.2.1)[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Remaining reads were mapped to the mm39 mouse genome using STAR (v2.7.10b) [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. BAM files were deduplicated (Sambamba, v0.8.2)[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] and filtered for high-quality alignments (SAMtools, v1.14)[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] with flags \u0026ldquo;-F 1804 -f 2 -q 20\u0026rdquo;.\u003c/p\u003e\u003cp\u003em\u0026sup1;A- and SRSF1-bound peaks were identified using exomePeak2 (v1.9.1)[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], which applies Wald tests on Poisson generalized linear models, with FDR correction via the Benjamini-Hochberg method. m\u0026sup1;A peak distribution along transcripts was visualized with Guitar (v2.10.0)[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Gene ontology (GO) analysis of m\u0026sup1;A-modified targets was performed with clusterProfiler (v3.18.1)[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], and enriched RNA motifs were identified using the MEME Suite (v5.5.2).\u003c/p\u003e\u003cp\u003eFor integrative analysis, overlap between m\u0026sup1;A-modified sites and SRSF1 peaks was computed using bedtools intersect (v2.30.0)[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Enrichment patterns were visualized via metagene plots using deepTools (v3.5.1)[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Correlations between m\u0026sup1;A and SRSF1 binding were assessed by binning the genome into 10 kb windows and quantifying read coverage with bedtools multicov followed by Pearson correlation analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eAll statistical analyses were performed using GraphPad Prism 9. Most quantitative results are presented as box-and-whisker plots displaying individual values, median, and range; bar graphs are shown as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM. For comparisons between two groups, data normality was first assessed using the Shapiro-Wilk test. If normality was not met, the nonparametric Mann-Whitney U test was used; otherwise, an unpaired two-tailed t-test was applied. Welch\u0026rsquo;s correction was used for t-tests when variances were unequal. For behavioral experiments, fear acquisition and extinction training data are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM, whereas retrieval test data are shown as box plots; all behavioral data were analyzed using two-way ANOVA followed by Tukey\u0026rsquo;s post hoc test for multiple group comparisons. Statistical significance was defined as follows: *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003em\u0026sup1;A accumulation in adult mPFC during extinction learning requires TRMT61A\u003c/h2\u003e\u003cp\u003eGiven that fear extinction serves as the neurobiological foundation of exposure therapy, we asked whether m\u0026sup1;A plays an active role in extinction itself. Mice underwent fear acquisition training followed by either extinction (EXT) training or retention control (RC) the next day (Supplementary Fig.\u0026nbsp;1). Mass spectrometry revealed a significant increase in total m\u0026sup1;A levels in the mPFC after EXT training (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). To investigate the potential regulator of m\u0026sup1;A accumulation during EXT, we examined known m\u0026sup1;A methyltransferase[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. As TRMT61B is not expressed in mice, analysis focused on the TRMT6/TRMT61A complex[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Although TRMT6 and TRMT61A mRNA levels remained unchanged following EXT training (supplementary Fig.\u0026nbsp;2), their protein levels were significantly increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB, C). Given that TRMT61A is the catalytic subunit of the TRMT6/TRMT61A m\u0026sup1;A methyltransferase complex, we next designed a lentiviral shRNA construct targeting TRMT61A and stereotaxically delivered it into the ILPFC (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF). Knockdown efficiency was validated in both primary cortical neurons (PCNs) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). After fear acquisition (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG), mice were divided to ensure comparable freezing scores within each group. ILPFC shTRMT61A knockdown reduced total m\u0026sup1;A level (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE) and did not affect extinction training process (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH), but significantly impaired extinction memory retrieval (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI). Open field performance was comparable, indicating no effect on locomotion or anxiety (Supplementary Fig.\u0026nbsp;3). Together, these findings indicate that TRMT61A-mediated m\u0026sup1;A methylation is dynamically regulated by extinction learning and is required for the proper formation of extinction memory.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003eExtinction reshapes the m\u0026sup1;A landscape toward synapses in the mPFC\u003c/h2\u003e\u003cp\u003eSince TRMT61A is required for extinction memory, we next mapped m\u0026sup1;A changes by m\u0026sup1;A RIP-seq in whole mPFC from RC and EXT mice. Motif analysis revealed conserved GA-rich consensus sequences in both groups, including a dominant AGGUAA motif (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Separate peak calling and overlap analysis identified 90 EXT-specific, 561 RC-specific, and 2,219 shared transcripts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Metagene profiles were broadly similar between RC and EXT across the 5\u0026prime;UTR, CDS, and 3\u0026prime;UTR (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Gene ontology (GO) analysis of m\u0026sup1;A-modified genes in RC highlighted synapse-related processes, whereas EXT retained synaptic terms and additionally enriched RNA-regulatory pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), suggesting a redistribution of m\u0026sup1;A targets with extinction at the synapse.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eMotivated by this shift, we fractionated mPFC into synaptic and nuclear compartments using Percoll-sucrose gradients, Western blots verified clean separation (synapse: PSD95⁺/H3⁻; nucleus: PSD95⁻/H3⁺) (Supplementary Fig.\u0026nbsp;4). Despite no change in TRMT6 and TRMT61A mRNA between RC and EXT (Supplementary Fig.\u0026nbsp;5), protein levels were selectively elevated in synaptic fractions after extinction, with no nuclear effect (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE-G). Consistently, Ribosome Immunoprecipitation (Ribo IP)-qPCR showed greater ribosome association of TRMT6 and TRMT61A mRNAs at synapses in EXT, with no change in nuclear fractions (Supplementary Fig.\u0026nbsp;6), supporting synapse-specific local translation that may contribute to local m\u0026sup1;A accrual and synaptic remodeling.\u003c/p\u003e\u003cp\u003eTo further characterize compartment-specific regulation, we performed m\u0026sup1;A RIP-seq in synaptic and nuclear fractions from RC and EXT mPFC. All four conditions (RC-nucleus, EXT-nucleus, RC-synapse, EXT-synapse) showed GA-enriched potential m\u0026sup1;A motifs consistent with the whole-tissue analysis, while the EXT-synapse group exhibited a distinct GGAAGA motif (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH). In nuclear fractions, m\u0026sup1;A remained enriched at the 5\u0026rsquo; UTR under both conditions, mirroring whole-tissue patterns; By contrast, extinction induced a redistribution of m\u0026sup1;A at synapses relative to nuclear and whole mPFC compartments, with reduced 5\u0026rsquo; UTR enrichment and increased CDS/3\u0026rsquo; UTR peaks (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eJ). GO terms overlapped across compartments (distal axon and myelin sheath), but nucleus-biased targets were preferentially associated with postsynaptic and dendritic spine-related processes, whereas synaptic m\u0026sup1;A-modified transcripts were selectively enriched for neurotransmitter transport, synaptic vesicle transport, and glutamatergic signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eK; Supplementary Fig.\u0026nbsp;7). A visual summary further highlighted increased synaptic 3\u0026prime;UTR-enriched peaks following extinction (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eL).\u003c/p\u003e\u003cp\u003eGiven the regulatory importance of the 3\u0026prime;UTR in post-transcriptional control[\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], we focused on synaptic transcripts with extinction-induced m\u0026sup1;A accumulation specifically in this region. Among 88 EXT-specific synaptic m\u0026sup1;A-modified transcripts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eI), \u003cem\u003ePten\u003c/em\u003e emerged as a prominent candidate. Extinction produced a narrow, high-intensity m\u0026sup1;A peak within the \u003cem\u003ePten\u003c/em\u003e 3\u0026prime;UTR at synapses, with no other peaks across the transcript and no signal in the nuclear compartment (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eM; Supplementary Fig.\u0026nbsp;8). This site-specific modification was validated by m\u0026sup1;A RIP-qPCR (Supplementary Fig.\u0026nbsp;9). Together, these findings demonstrate compartment- and region-specific reshaping of m\u0026sup1;A during extinction learning, with a synaptic bias coupled to increased TRMT6/61A at synapses.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSynaptic knockdown of\u003c/b\u003e \u003cb\u003ePten\u003c/b\u003e \u003cb\u003eimpairs fear extinction memory\u003c/b\u003e\u003c/p\u003e\u003cp\u003eProtein analysis revealed that extinction training significantly increased PTEN expression in both total and synaptic mPFC fractions, whereas nuclear \u003cem\u003ePten\u003c/em\u003e levels remained unchanged (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-C), implicating a synapse-specific regulatory mechanism. To directly assess the functional relevance of synaptic \u003cem\u003ePten\u003c/em\u003e, we employed a CRISPR-Cas-Inspired RNA Targeting System (CIRTS)-based RNA knockdown approach incorporating the Calm3 intron, which recruits RNA binding protein Staufen2 to enable selective targeting of synaptic transcripts[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Using a guide RNA specific to \u003cem\u003ePten\u003c/em\u003e mRNA, this system delivered the PIN RNase to selectively degrade \u003cem\u003ePten\u003c/em\u003e mRNA at synapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). CIRTS system selectively reduced synaptic \u003cem\u003ePten\u003c/em\u003e mRNA in both the ILPFC and PCNs, with minimal impact on nuclear \u003cem\u003ePten\u003c/em\u003e levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, F). Functionally, as outlined in the experimental workflow (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG), synaptic knockdown of \u003cem\u003ePten\u003c/em\u003e impaired extinction learning, reflected by altered extinction curves and a specific post hoc difference at the 9th CS presentation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH). In contrast, baseline freezing in a novel context without extinction was unaffected (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI, PreCS), indicating that the effect was specific to extinction learning rather than baseline freezing. Notably, knockdown mice exhibited significantly increased freezing 24 hours after training (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI), indicating impaired retrieval and confirming a disruption in extinction memory consolidation. To exclude potential confounds, we next evaluated anxiety and locomotion. In the open field test, knockdown had no effect on center time, center entries, or total distance traveled (Supplementary Fig.\u0026nbsp;10A-C). In the elevated plus maze, open arm entries and time spent in open arms were similarly unaffected (Supplementary Fig.\u0026nbsp;10D, E). These findings support a specific role for synaptic \u003cem\u003ePten\u003c/em\u003e in extinction memory, rather than general changes in anxiety or locomotor activity levels. Together, these findings identify synaptic PTEN upregulation as a critical component of extinction memory formation, raising the possibility that m\u0026sup1;A may contribute to this process by modulating \u003cem\u003ePten\u003c/em\u003e at the post-transcriptional level.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003em\u0026sup1;A stabilizes synaptic\u003c/b\u003e \u003cb\u003ePten\u003c/b\u003e \u003cb\u003emRNA, promoting its protein accumulation and extinction memory\u003c/b\u003e\u003c/p\u003e\u003cp\u003eGiven that synaptic \u003cem\u003ePten\u003c/em\u003e is required for both the acquisition and retrieval of extinction memory, we next investigated the underlying mechanism. Specifically, we asked whether m\u0026sup1;A methylation at the 3\u0026rsquo; UTR of \u003cem\u003ePten\u003c/em\u003e mRNA contributes to this regulation by promoting transcript stability and facilitating local protein synthesis during extinction. In PCNs, TRMT61A knockdown significantly reduced \u003cem\u003ePten\u003c/em\u003e mRNA stability under depolarizing (20 mM KCl) conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), suggesting that m\u0026sup1;A promotes \u003cem\u003ePten\u003c/em\u003e stability under activity-dependent conditions. Consistently, m\u0026sup1;A RIP-qPCR revealed a marked decrease in m\u0026sup1;A enrichment at the \u003cem\u003ePten\u003c/em\u003e 3\u0026rsquo; UTR in TRMT61A-deficient neurons and ILPFC (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), confirming that this modification is TRMT61A-dependent. These findings indicate that m\u0026sup1;A enhances \u003cem\u003ePten\u003c/em\u003e mRNA stability, which promote its protein accumulation during extinction retrieval.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo assess the functional relevance of this m\u0026sup1;A event in vivo, we engineered a synapse-specific demethylation system by fusing ALKBH3 to the CIRTS-Calm3 system (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). This construct selectively removed m\u0026sup1;A from the 3\u0026rsquo; UTR of \u003cem\u003ePten\u003c/em\u003e mRNA in synaptic compartments of the ILPFC, with no detectable effect in nuclear fractions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). As outlined in the experimental workflow (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), synaptic depletion of \u003cem\u003ePten\u003c/em\u003e m\u0026sup1;A did not impair extinction learning (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF), but disrupted extinction memory retrieval (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). Baseline freezing behavior (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG, PreCS) and exploratory activity (Supplementary Fig.\u0026nbsp;11) remained unaffected, partially resembling the behavioral phenotype observed following synaptic \u003cem\u003ePten\u003c/em\u003e knockdown. Synaptic depletion of \u003cem\u003ePten\u003c/em\u003e m\u0026sup1;A impaired extinction retrieval, with anxiety-like behavior and locomotion unchanged. The convergent behavioral deficits after synaptic \u003cem\u003ePten\u003c/em\u003e knockdown and targeted m\u0026sup1;A demethylation thus support a synapse-localized m\u0026sup1;A to \u003cem\u003ePten\u003c/em\u003e mechanism in extinction memory.\u003c/p\u003e\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\u003ch2\u003eExtinction learning selectively recruits synaptic m\u0026sup1;A readers including SRSF1\u003c/h2\u003e\u003cp\u003eRNA modifications can regulate gene expression via recruitment of RBPs. To identify RBPs recognizing m\u0026sup1;A-modified transcripts during extinction, we performed pull-downs with synthetic oligos containing a canonical m\u0026sup1;A motif from RIP-seq and prior reports[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Matched unmodified oligos served as controls, and bound proteins were analyzed by mass spectrometry (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). We applied stringent criteria across biological replicates-consistent detection in EXT-synapse m\u0026sup1;A pull-downs, preferential enrichment over A-oligos in EXT but not RC conditions, and moderate log₂(m\u0026sup1;A/A) fold enrichment (1\u0026ndash;8), identified six RBPs: SRSF1, MYEF2, ALYREF, FBL, DDX28, and RPL3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). These proteins have been associated with diverse neuronal processes, including synaptic RNA metabolism[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], myelination[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e], mRNA export[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e], and activity-dependent translation[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Their selective enrichment in EXT-synapse fractions indicates extinction-induced recruitment of neuronally relevant m\u0026sup1;A readers (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Notably, unlike canonical readers YTHDF1-3 and YTHDC1, which were absent from synaptic fractions and not enriched in nuclear fractions (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), extinction recruits a distinct set of synapse-localized m\u0026sup1;A readers.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next aimed to define the m\u0026sup1;A reader that binds \u003cem\u003ePten\u003c/em\u003e\u0026rsquo; s 3\u0026rsquo; UTR and could couple methylation to function. Using RBPmap database, we screened the sequence surrounding the \u003cem\u003ePten\u003c/em\u003e synaptic m\u0026sup1;A peak and identified 93 candidate RBPs. Cross-referencing with the literature narrowed this to CPEB1, IGF2BP1, IGF2BP2, MSI1, RBM24, RBM38, and SRSF1protein, all previously reported to interact with \u003cem\u003ePten\u003c/em\u003e mRNA. SRSF1 uniquely overlapped both prediction and MS-based identification, implicating it as a likely m\u0026sup1;A-dependent interactor. We performed fRIP-qPCR and fRIP-seq to confirm these interactions (workflow in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). fRIP-qPCR showed that extinction enhanced SRSF1 binding to the \u003cem\u003ePten\u003c/em\u003e 3\u0026rsquo; UTR in synaptic, but not nuclear fractions of mPFC (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). This interaction was reduced by CIRTS-ALKBH3 demethylation in PCNs (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG), confirming m\u0026sup1;A-dependence. To assess broader association, we performed SRSF1 fRIP-seq in whole mPFC and compared peaks with m\u0026sup1;A RIP-seq.\u0026nbsp;The substantial overlap of SRSF1-bound RNAs with m\u0026sup1;A-modified transcripts (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH) further supports a broader role for SRSF1 in recognizing m\u0026sup1;A-modified RNAs.\u003c/p\u003e\u003cp\u003eTogether, these findings identify SRSF1 as an extinction-induced, synapse-enriched m\u0026sup1;A reader that binds \u003cem\u003ePten\u003c/em\u003e in an m\u0026sup1;A-dependent manner and may mediate broader post-transcriptional regulation during fear extinction.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSynaptic m\u0026sup1;A depletion of\u003c/b\u003e \u003cb\u003ePten\u003c/b\u003e \u003cb\u003eimpairs dendritic and synaptic structure\u003c/b\u003e\u003c/p\u003e\u003cp\u003em\u0026sup1;A at the \u003cem\u003ePten\u003c/em\u003e 3\u0026rsquo; UTR stabilizes the transcript and promote its protein accumulation during extinction, contributing to fear extinction memory, but whether it also drives synaptic structural remodeling remains unclear. In parallel, prior evidence implicating \u003cem\u003ePten\u003c/em\u003e in regulating synaptic architecture[\u003cspan additionalcitationids=\"CR44 CR45\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e], prompting us to investigate whether synaptic depletion of m\u0026sup1;A at the 3\u0026rsquo;UTR of \u003cem\u003ePten\u003c/em\u003e leads to comparable morphological alterations. Using the CIRTS-ALKBH3 system to selectively demethylate the 3\u0026rsquo; UTR of \u003cem\u003ePten\u003c/em\u003e mRNA at synapses, we assessed dendritic architecture in the ILPFC following extinction training. Golgi staining revealed a marked reduction in total dendritic length and branch complexity in mice with synaptic m\u0026sup1;A depletion compared to controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-C). Dendritic spine analysis further revealed a significant reduction in spine density (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD-E), suggesting compromised structural plasticity. Given the observed dendritic deficits and the synaptic enrichment of vesicle-related transcripts identified in our m\u0026sup1;A RIP-seq data, we next examined presynaptic ultrastructure via transmission electron microscopy (TEM). Synaptic m\u0026sup1;A depletion of \u003cem\u003ePten\u003c/em\u003e reduced postsynaptic density (PSD) thickness, synaptic vesicle (SV) density, and increased cleft width, with no change in active zone (AZ) length (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH-L). Consistently, levels of synaptic proteins PSD95 and SYN1 were significantly reduced in ILPFC lysates from the m\u0026sup1;A-depleted group (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF-G). Together, these findings demonstrate that extinction-induced m\u003csup\u003e1\u003c/sup\u003eA modification of \u003cem\u003ePten\u003c/em\u003e mRNA at synapses supports not only behavioral plasticity but also the structural integrity of dendrites and synapses. This provides a mechanistic link between localized epitranscriptomic regulation and the anatomical remodeling required for extinction memory formation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003em\u0026sup1;A is functionally implicated in diverse cellular processes, including tumor progression, stress responses, and oxidative adaptation[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan additionalcitationids=\"CR48 CR49 CR50 CR51\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. However, its role in spatially defined RNA regulation within the adult brain, particularly at synapses relevant for learning and memory, remains uncharacterized. Although synaptic m⁶A accumulation has been linked to fear extinction[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], the potential involvement of m\u0026sup1;A in this form of plasticity is still unknown. We uncover a synapse-specific regulatory mechanism in which extinction training triggers TRMT6 and TRMT61A accumulation at synapses, leading to redistribution of m\u0026sup1;A marks and selective 3\u0026rsquo; UTR methylation of \u003cem\u003ePten\u003c/em\u003e mRNA. This m\u0026sup1;A modification promotes \u003cem\u003ePten\u003c/em\u003e transcript stability and enhances protein expression through recruitment of SRSF1, identified here as a synaptic m\u0026sup1;A reader. These results define a mechanism in which TRMT61A installs m\u0026sup1;A to promote SRSF1 binding on \u003cem\u003ePten\u003c/em\u003e mRNA, thereby translating experience into molecular signals governing synaptic plasticity.\u003c/p\u003e\u003cp\u003eOur findings suggest that m\u0026sup1;A is dynamically regulated by experience and may contribute to memory flexibility impairments observed in PTSD. In mice, extinction training elevated both m\u0026sup1;A abundance and TRMT6/61A expression in the mPFC, whereas knockdown of TRMT61A in the ILPFC impaired extinction memory, highlighting a functional role of m\u0026sup1;A in behavioral adaptation. At the transcriptomic level, extinction was accompanied by widespread remodeling of m\u0026sup1;A marks in the mPFC, particularly on genes related to synaptic organization and signaling. Together, these results suggest that m\u0026sup1;A promotes extinction by tuning synaptic gene expression networks. Motivated by this, we observed synapse-specific upregulation of TRMT6 and TRMT61A after extinction, which led us to investigate whether m\u0026sup1;A distribution similarly exhibits compartmental specificity. Subsequent subcellular m\u0026sup1;A RIP-seq revealed that synaptic m\u0026sup1;A preferentially enriched at coding sequences and 3\u0026rsquo; UTRs, a compartment-specific pattern absents in nuclear or whole-tissue profiles. Together, these findings suggest that extinction induces synapse-specific m\u0026sup1;A remodeling to support local post-transcriptional regulation essential for fear extinction, prompting us to identify the relevant downstream effectors.\u003c/p\u003e\u003cp\u003eBuilding on the role of 3\u0026rsquo; UTRs in post-transcriptional regulation[\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], we focused on synaptic transcripts with extinction-induced m\u0026sup1;A enrichment in this region. Among these, \u003cem\u003ePten\u003c/em\u003e mRNA was particularly notable, consistent with its known involvement in synaptic plasticity and cognitive function[\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Using a CIRTS-PIN-Calm3 system, we performed synapse-specific knockdown of \u003cem\u003ePten\u003c/em\u003e in the ILPFC, which impaired extinction memory formation. This demonstrates the necessity of synaptic \u003cem\u003ePten\u003c/em\u003e for behavioral plasticity.\u003c/p\u003e\u003cp\u003eInterestingly, our prior work in the basolateral amygdala (BLA) suggested that PTEN negatively regulates extinction[\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e], in apparent contrast to our observations in the ILPFC. This divergence likely reflects region- and circuit-level specificity in extinction control by PTEN. It may also arise from cell-type-specific effects: PTEN restricts dendritic growth in excitatory neurons, whereas its loss in interneurons accelerates maturation and limits plasticity via increased perineuronal nets[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. In addition, manipulation scale matters: systemic inhibition of the PI3K-mTOR pathway impairs extinction[\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], but such broad interventions differ fundamentally from our synapse-specific perturbation, which selectively disrupted extinction acquisition, highlighting the importance of spatial and cellular precision when modulating PTEN function. Finally, developmental stage further shapes outcomes: while \u003cem\u003ePten\u003c/em\u003e loss promotes synaptogenesis in juveniles [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e], in our adult IL circuits \u003cem\u003ePten\u003c/em\u003e stabilizes synapses. We therefore propose that, in the mature ILPFC, PTEN supports extinction by maintaining synaptic integrity, thereby contributing to the regulation of extinction.\u003c/p\u003e\u003cp\u003eTo determine whether m\u0026sup1;A directly regulates synaptic \u003cem\u003ePten\u003c/em\u003e, we used the CIRTS-ALKBH3-Calm3 system to selectively demethylate its 3\u0026rsquo; UTR. This site-specific removal of m\u0026sup1;A recapitulated the behavioral impairment observed with synaptic \u003cem\u003ePten\u003c/em\u003e knockdown. Since RNA modifications modulate transcript fate by affecting stability and translation[\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e], we further examined this mechanism. Demethylation reduced \u003cem\u003ePten\u003c/em\u003e mRNA stability in neurons, suggesting that m\u0026sup1;A enhances transcript longevity and supports protein expression, which is consistent with extinction-induced upregulation of synaptic PTEN protein.\u003c/p\u003e\u003cp\u003eRNA modifications are known to function via recruitment of RNA-binding proteins (RBPs)[\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. In line with this, the synaptic enrichment of m\u0026sup1;A observed here may reflect selective engagement of RBPs during extinction learning. Using synthetic probes containing m\u0026sup1;A motifs, we performed RNA pull-down assays and identified six candidate RBPs enriched in the EXT-synapse condition. Among these, SRSF1 emerged as a top hit. Notably, previous studies have shown that SRSF1 binds the 3' UTR of \u003cem\u003ePten\u003c/em\u003e mRNA and enhances its stability and translation[\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e], consistent with our findings that m\u0026sup1;A promotes synaptic PTEN expression. We further demonstrated that in EXT process, this interaction is disrupted by m\u0026sup1;A demethylation, confirming that SRSF1 selectively engages m\u0026sup1;A-modified transcripts in a methylation-dependent manner. Importantly, whole-mPFC SRSF1 fRIP-seq revealed substantial overlap between SRSF1-bound regions and m\u0026sup1;A-modified transcripts in extinction-trained mice, indicating that SRSF1 functions not only as a noncanonical m\u0026sup1;A reader, but also as a core effector of synaptic RNA methylation in behaviorally relevant neural circuits. Together, these findings position SRSF1 as a molecular interpreter of synaptic m\u0026sup1;A marks, translating RNA methylation signals into transcript stabilization and local protein synthesis. This reveals a critical role for SRSF1 in bridging epitranscriptomic regulation with activity-dependent synaptic plasticity.\u003c/p\u003e\u003cp\u003eGiven prior evidence implicating PTEN in regulating synaptic architecture[\u003cspan additionalcitationids=\"CR44 CR45\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e], our results demonstrate that extinction-induced, m\u0026sup1;A-dependent stabilization of \u003cem\u003ePten\u003c/em\u003e mRNA contributes to structural plasticity in the ILPFC. Selective demethylation of the m\u0026sup1;A site at the 3\u0026rsquo; UTR of \u003cem\u003ePten\u003c/em\u003e led to pronounced impairments in dendritic morphology, including reduced arbor length, diminished branching complexity, and decreased spine density. Transmission electron microscopy further revealed attenuated PSD thickness, reduced synaptic vesicle density, and widened synaptic clefts, hallmarks of compromised synaptic ultrastructure, alongside decreased expression of core synaptic proteins such as SYN1 and PSD95. These data demonstrate that m\u0026sup1;A-mediated regulation of \u003cem\u003ePten\u003c/em\u003e is essential for maintaining synaptic integrity in extinction-relevant cortical circuits. By linking epitranscriptomic control to structural remodeling, this mechanism provides a molecular substrate through which experience modulates synaptic architecture to enable adaptive memory formation.\u003c/p\u003e\u003cp\u003eIn summary, our findings delineate a synapse-specific TRMT61A-dependent m\u0026sup1;A methylation at synapses promotes SRSF1 recruitment to \u003cem\u003ePten\u003c/em\u003e mRNA, links RNA methylation to local transcript stabilization, protein synthesis, and extinction memory formation. Through extinction-induced redistribution of m\u0026sup1;A and selective 3\u0026rsquo; UTR methylation of \u003cem\u003ePten\u003c/em\u003e mRNA, this pathway enhances synaptic PTEN expression via SRSF1 recruitment, supporting structural and behavioral plasticity. These results identify synaptic m\u0026sup1;A remodeling as a mechanistic contributor to memory flexibility and suggest its dysregulation may underlie cognitive impairments in disorders such as PTSD. In line with this possibility, plasma m\u0026sup1;A levels were higher in healthy controls (HC) than in PTSD patients (Supplementary Fig.\u0026nbsp;12), paralleling the trend seen in extinction-trained versus retrieval control mice. By uncovering a spatially precise layer of post-transcriptional control, this work provides a framework for future studies of RNA-based mechanisms in experience-dependent neural adaptation.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eConflict of Interest\u003c/h2\u003e\u003cp\u003eThe authors declare no competing financial interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eThe authors are grateful to Prof. Timothy W. Bredy (University of Queensland) for kindly providing the CIRTS-PIN-Calm3 plasmid. This study was supported by the National Natural Science Foundation of China (Grant No. 82171517; 82271556; 82471534), the Major State Research Development Program of Hubei (Grant No. 2024BCA003) and Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province (No. 2023ZQNZD016).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eKessler RC, Aguilar-Gaxiola S, Alonso J, Benjet C, Bromet EJ, Cardoso G, et al. Trauma and PTSD in the WHO World Mental Health Surveys. Eur J Psychotraumatol. 2017;8(sup5):1353383.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoenen KC, Ratanatharathorn A, Ng L, McLaughlin KA, Bromet EJ, Stein DJ, et al. Posttraumatic stress disorder in the World Mental Health Surveys. Psychol Med. 2017;47(13):2260\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCharlson FJ, Flaxman A, Ferrari AJ, Vos T, Steel Z, Whiteford HA. Post-traumatic stress disorder and major depression in conflict-affected populations: an epidemiological model and predictor analysis. Glob Ment Health (Camb). 2016;3:e4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFlores A, Fullana MA, Soriano-Mas C, Andero R. Lost in translation: how to upgrade fear memory research. Mol Psychiatry. 2018;23(11):2122\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRessler KJ, Berretta S, Bolshakov VY, Rosso IM, Meloni EG, Rauch SL, et al. Post-traumatic stress disorder: clinical and translational neuroscience from cells to circuits. Nat Rev Neurol. 2022;18(5):273\u0026ndash;88.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeterson AL, Blount TH, Foa EB, Brown LA, McLean CP, Mintz J, et al. Massed vs Intensive Outpatient Prolonged Exposure for Combat-Related Posttraumatic Stress Disorder: A Randomized Clinical Trial. JAMA Netw Open. 2023;6(1):e2249422.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMilad MR, Quirk GJ. Neurons in medial prefrontal cortex signal memory for fear extinction. Nature. 2002;420(6911):70\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun C, Nold A, Fusco CM, Rangaraju V, Tchumatchenko T, Heilemann M, et al. The prevalence and specificity of local protein synthesis during neuronal synaptic plasticity. Sci Adv. 2021;7(38):eabj0790.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Oostrum M, Schuman EM. Understanding the molecular diversity of synapses. Nat Rev Neurosci. 2025;26(2):65\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNainar S, Marshall PR, Tyler CR, Spitale RC, Bredy TW. Evolving insights into RNA modifications and their functional diversity in the brain. Nat Neurosci. 2016;19(10):1292\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMadugalle SU, Liau WS, Zhao Q, Li X, Gong H, Marshall PR, et al. Synapse-Enriched m(6)A-Modified Malat1 Interacts with the Novel m(6)A Reader, DPYSL2, and Is Required for Fear-Extinction Memory. J Neurosci. 2023;43(43):7084\u0026ndash;100.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang C, Jia G. Reversible RNA Modification N(1)-methyladenosine (m(1)A) in mRNA and tRNA. Genomics Proteomics Bioinformatics. 2018;16(3):155\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Xiong X, Zhang M, Wang K, Chen Y, Zhou J, et al. Base-Resolution Mapping Reveals Distinct m(1)A Methylome in Nuclear- and Mitochondrial-Encoded Transcripts. Mol Cell. 2017;68(5):993\u0026ndash;1005 e9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAnderson J, Phan L, Hinnebusch AG. The Gcd10p/Gcd14p complex is the essential two-subunit tRNA(1-methyladenosine) methyltransferase of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A. 2000;97(10):5173\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen M, Chen Y, Wang K, Deng X, Chen J. [Not Available]. Clin Transl Med. 2024;14(6):e1666.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Xiong X, Wang K, Wang L, Shu X, Ma S, et al. Transcriptome-wide mapping reveals reversible and dynamic N(1)-methyladenosine methylome. Nat Chem Biol. 2016;12(5):311\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDominissini D, Nachtergaele S, Moshitch-Moshkovitz S, Peer E, Kol N, Ben-Haim MS, et al. The dynamic N(1)-methyladenosine methylome in eukaryotic messenger RNA. Nature. 2016;530(7591):441\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSafra M, Sas-Chen A, Nir R, Winkler R, Nachshon A, Bar-Yaacov D, et al. The m1A landscape on cytosolic and mitochondrial mRNA at single-base resolution. Nature. 2017;551(7679):251\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang C, Yi X, Hou M, Li Q, Li X, Lu L, et al. The landscape of m(1)A modification and its posttranscriptional regulatory functions in primary neurons. Elife. 2023;12.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQi Z, Zhang C, Jian H, Hou M, Lou Y, Kang Y, et al. N(1)-Methyladenosine modification of mRNA regulates neuronal gene expression and oxygen glucose deprivation/reoxygenation induction. Cell Death Discov. 2023;9(1):159.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiau WS, Zhao Q, Bademosi A, Gormal RS, Gong H, Marshall PR, et al. Fear extinction is regulated by the activity of long noncoding RNAs at the synapse. Nat Commun. 2023;14(1):7616.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWei W, Zhao Q, Wang Z, Liau WS, Basic D, Ren H, et al. ADRAM is an experience-dependent long noncoding RNA that drives fear extinction through a direct interaction with the chaperone protein 14-3-3. Cell Rep. 2022;38(12):110546.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDunkley PR, Jarvie PE, Robinson PJ. A rapid Percoll gradient procedure for preparation of synaptosomes. Nat Protoc. 2008;3(11):1718\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSelth LA, Gilbert C, Svejstrup JQ. RNA immunoprecipitation to determine RNA-protein associations in vivo. Cold Spring Harb Protoc. 2009;2009(6):pdb prot5234.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD GH, Kelley DR, Tenen D, Bernstein B, Rinn JL. Widespread RNA binding by chromatin-associated proteins. Genome Biol. 2016;17:28.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884-i90.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang Y, Park C, Bennett C, Thornton M, Kim D. Rapid and accurate alignment of nucleotide conversion sequencing reads with HISAT-3N. Genome Res. 2021;31(7):1290\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTarasov A, Vilella AJ, Cuppen E, Nijman IJ, Prins P. Sambamba: fast processing of NGS alignment formats. Bioinformatics. 2015;31(12):2032\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDanecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMeng J, Lu Z, Liu H, Zhang L, Zhang S, Chen Y, et al. A protocol for RNA methylation differential analysis with MeRIP-Seq data and exomePeak R/Bioconductor package. Methods. 2014;69(3):274\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCui X, Wei Z, Zhang L, Liu H, Sun L, Zhang SW, et al. Guitar: An R/Bioconductor Package for Gene Annotation Guided Transcriptomic Analysis of RNA-Related Genomic Features. Biomed Res Int. 2016;2016:8367534.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3):100141.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQuinlan AR. BEDTools: The Swiss-Army Tool for Genome Feature Analysis. Curr Protoc Bioinformatics. 2014;47:11 2 1\u0026ndash;34.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamirez F, Dundar F, Diehl S, Gruning BA, Manke T. deepTools: a flexible platform for exploring deep-sequencing data. Nucleic Acids Res. 2014;42(Web Server issue):W187-91.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBushkin GG, Pincus D, Morgan JT, Richardson K, Lewis C, Chan SH, et al. m(6)A modification of a 3' UTR site reduces RME1 mRNA levels to promote meiosis. Nat Commun. 2019;10(1):3414.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatoulkova E, Michalova E, Vojtesek B, Hrstka R. The role of the 3' untranslated region in post-transcriptional regulation of protein expression in mammalian cells. RNA Biol. 2012;9(5):563\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang X, Zhao BS, Roundtree IA, Lu Z, Han D, Ma H, et al. N(6)-methyladenosine Modulates Messenger RNA Translation Efficiency. Cell. 2015;161(6):1388\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHulse RP, Drake RA, Bates DO, Donaldson LF. The control of alternative splicing by SRSF1 in myelinated afferents contributes to the development of neuropathic pain. Neurobiol Dis. 2016;96:186\u0026ndash;200.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Y, Zhang Y, Zhang S, Kim B, Hull VL, Xu J, et al. PARP1-mediated PARylation activity is essential for oligodendroglial differentiation and CNS myelination. Cell Rep. 2021;37(1):109695.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShi M, Zhang H, Wu X, He Z, Wang L, Yin S, et al. ALYREF mainly binds to the 5' and the 3' regions of the mRNA in vivo. Nucleic Acids Res. 2017;45(16):9640\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiona F, Beretta S, Zippo A, Stefanoni A, Tomasoni Z, Vicidomini C, et al. Shank3 modulates Rpl3 expression and protein synthesis via mGlu5: implications for Phelan McDermid syndrome. Mol Psychiatry. 2025;30(8):3599\u0026ndash;614.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKnafo S, Sanchez-Puelles C, Palomer E, Delgado I, Draffin JE, Mingo J, et al. PTEN recruitment controls synaptic and cognitive function in Alzheimer's models. Nat Neurosci. 2016;19(3):443\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKwon CH, Luikart BW, Powell CM, Zhou J, Matheny SA, Zhang W, et al. Pten regulates neuronal arborization and social interaction in mice. Neuron. 2006;50(3):377\u0026ndash;88.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTariq K, Cullen E, Getz SA, Conching AKS, Goyette AR, Prina ML, et al. Disruption of mTORC1 rescues neuronal overgrowth and synapse function dysregulated by Pten loss. Cell Rep. 2022;41(5):111574.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD'Amore A, Sundberg M, Lin R, Lubbers ET, Winden KD, Yu L, et al. Phenotypic rescue via mTOR inhibition in neuron-specific Pten knockout mice reveals AKT and mTORC1-site specific changes. Mol Psychiatry. 2025;30(7):3077\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUeda Y, Ooshio I, Fusamae Y, Kitae K, Kawaguchi M, Jingushi K, et al. AlkB homolog 3-mediated tRNA demethylation promotes protein synthesis in cancer cells. Sci Rep. 2017;7:42271.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Y, Wang J, Li X, Xiong X, Wang J, Zhou Z, et al. N(1)-methyladenosine methylation in tRNA drives liver tumourigenesis by regulating cholesterol metabolism. Nat Commun. 2021;12(1):6314.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWoo HH, Chambers SK. Human ALKBH3-induced m(1)A demethylation increases the CSF-1 mRNA stability in breast and ovarian cancer cells. Biochim Biophys Acta Gene Regul Mech. 2019;1862(1):35\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamato I, Sho M, Shimada K, Hotta K, Ueda Y, Yasuda S, et al. PCA-1/ALKBH3 contributes to pancreatic cancer by supporting apoptotic resistance and angiogenesis. Cancer Res. 2012;72(18):4829\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu Y, Chen Z, Xie G, Zhang H, Wang Z, Zhou J, et al. RNA m(1)A methylation regulates glycolysis of cancer cells through modulating ATP5D. Proc Natl Acad Sci U S A. 2022;119(28):e2119038119.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlriquet M, Calloni G, Martinez-Limon A, Delli Ponti R, Hanspach G, Hengesbach M, et al. The protective role of m1A during stress-induced granulation. J Mol Cell Biol. 2021;12(11):870\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKnafo S, Esteban JA. PTEN: Local and Global Modulation of Neuronal Function in Health and Disease. Trends Neurosci. 2017;40(2):83\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChwin N, Kiryk A. The integral role of PTEN in brain function: from neurogenesis to synaptic plasticity and social behavior. Acta Neurobiol Exp (Wars). 2025;84(4):309\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMurphy CP, Li X, Maurer V, Oberhauser M, Gstir R, Wearick-Silva LE, et al. MicroRNA-Mediated Rescue of Fear Extinction Memory by miR-144-3p in Extinction-Impaired Mice. Biol Psychiatry. 2017;81(12):979\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKritman M, Maroun M. Inhibition of the PI3 kinase cascade in corticolimbic circuit: temporal and differential effects on contextual fear and extinction. Int J Neuropsychopharmacol. 2013;16(4):825\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerandones C, Costanzo RV, Kowaljow V, Pivetta OH, Carminatti H, Radrizzani M. Correlation between synaptogenesis and the PTEN phosphatase expression in dendrites during postnatal brain development. Brain Res Mol Brain Res. 2004;128(1):8\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMeyer KD, Jaffrey SR. The dynamic epitranscriptome: N6-methyladenosine and gene expression control. Nat Rev Mol Cell Biol. 2014;15(5):313\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoundtree IA, Evans ME, Pan T, He C. Dynamic RNA Modifications in Gene Expression Regulation. Cell. 2017;169(7):1187\u0026ndash;200.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKatsuyama T, Li H, Comte D, Tsokos GC, Moulton VR. Splicing factor SRSF1 controls T cell hyperactivity and systemic autoimmunity. J Clin Invest. 2019;129(12):5411\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7703002/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7703002/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePost-traumatic stress disorder (PTSD) is marked by intrusive fear memories and impaired extinction learning, yet the underlying molecular mechanism at synapses remains elusive. Here, we uncover a previously unrecognized epitranscriptomic mechanism linking RNA modification to adaptive memory. We show that extinction learning increases synaptic TRMT6/TRMT61A levels, driving the redistribution of m¹A toward activity-relevant transcripts. Explicitly, using subcellular fractionation combined with m¹A RIP-seq, we identify a synapse-specific m¹A pattern, distinct from nuclear and whole mPFC profiles, marked by reduced 5’ UTR and enriched coding region and 3’ UTR methylation during extinction learning. Moreover, TRMT61A-mediated m¹A deposition in the 3’ UTR of \u003cem\u003ePten \u003c/em\u003emRNA within synaptic compartments enhances transcript stability and thereby increases PTEN protein levels. RNA pull-down coupled with mass spectrometry identified SRSF1 as a synapse-enriched, extinction-responsive RNA binding protein (RBP) that specifically binds m¹A and mediates downstream post-transcriptional regulation. Notably, TRMT61A-mediated m¹A modification and its recognition by SRSF1 converge on \u003cem\u003ePten\u003c/em\u003e regulation to drive synaptic remodeling and facilitate extinction memory. Site-specific disrupting \u003cem\u003ePten\u003c/em\u003e m¹A site impairs dendritic structure and blocks extinction retrieval, completing a mechanistic loop from RNA mark to behavior. Our findings highlight m¹A as a dynamic regulator of synaptic plasticity and identify a novel molecular circuit that may be targeted for PTSD therapeutics.\u003c/p\u003e","manuscriptTitle":"Synaptic m1A remodeling in the mPFC promotes fear extinction via Pten regulation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-03 07:40:17","doi":"10.21203/rs.3.rs-7703002/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"626ff402-23d4-47eb-a381-1838a351a764","owner":[],"postedDate":"October 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55636174,"name":"Biological sciences/Molecular biology"},{"id":55636175,"name":"Biological sciences/Neuroscience"},{"id":55636176,"name":"Health sciences/Diseases/Psychiatric disorders"},{"id":55636177,"name":"Biological sciences/Genetics"},{"id":55636178,"name":"Biological sciences/Psychology"}],"tags":[],"updatedAt":"2025-10-03T07:40:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-03 07:40:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7703002","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7703002","identity":"rs-7703002","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}