Identification of moonlighting, extranuclear and syndapin I-related functions of EPOP in mature neurons

Identification of moonlighting, extranuclear and syndapin I-related functions of EPOP in mature neurons | 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 Identification of moonlighting, extranuclear and syndapin I-related functions of EPOP in mature neurons Michael Kessels, Jacqueline Dömming, Regina Dahlhaus, Nicole Koch, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9117199/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Revealing functions of protein components distinct from initially described context significantly advances scientific knowledge. Here, we describe functions of EPOP (Elongin BC and Polycomb repressive complex 2-associated protein) distinct from EPOP’s established involvement in epigenetic regulation of gene expression in stem cells. These moonlighting functions that take place in a different cellular compartment, with a different binding partner and in mature, postmitotic cells. In neurons, we identified an EPOP subpool that was not nuclear but cytoplasmic and interacted with the membrane-shaping protein syndapin I. EPOP hereby showed a preference for the somatodendritic compartment and was also present in dendritic spines. Syndapin I KO neurons revealed that EPOP’s extranuclear presence depended on its direct binding partner syndapin I. EPOP loss-of-function negatively impacted the density, morphology and postsynaptic-density composition of mushroom-type dendritic spines in mature hippocampal neurons and by closely phenocopying syndapin I loss-of-function phenotypes highlighted an intimate functional relationship of both components in postsynapses. Biological sciences/Neuroscience/Cellular neuroscience Biological sciences/Neuroscience/Development of the nervous system/Synaptic development Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction The polycomb repressive complex 2 (PRC2) represents an important chromatin regulatory complex and functions by repressing the transcription of diverse developmental genes. PRC2 consists of a core complex responsible for catalyzing both di- and trimethylation of histone H3 at lysine 27 and various associated components that modulate its function in different ways (Fischer et al., 2022; Aguilar et al., 2025). In the nervous system, PRC2 has been reported to play important roles in neuronal identity, proliferation and differentiation of neural stem and progenitor cells as well as in gliogenesis. Mutations or dysregulations of PRC2 components are frequently associated with neurological diseases (Liu et al., 2018). EPOP (Elongin BC and Polycomb repressive complex 2-associated protein; also named C17orf96, esPRC2p48, E130012A19Rik) is a mammalian-specific PRC2-associated factor (Zhang et al., 2011; De Cegli et al., 2013; Smits et al., 2013; Alekseyenko et al., 2014). Whereas EPOP was shown to enhance the methyltransferase activity of PRC2 in vitro (Behringer et al., 2016; Zhang et al., 2011), EPOP deletion resulted in an elevated chromatin association of PRC2 and increased repression of PRC2 target genes suggesting that EPOP represses PRC2 function in vivo (Liefke & Shi, 2015; Behringer et al., 2016; Liefke et al., 2016; Healy et al., 2019; Granat et al., 2026). Recently, EPOP was shown to restrict the targeting of specifically the PCR2.1 complex to chromatin by disrupting the dimeric architecture of the enzyme complex (Gong et al., 2026). EPOP furthermore also interacts with the heterodimer Elongin BC (Behringer et al., 2016; Liefke et al., 2016), which plays a role in transcription elongation and protein turnover (Okumara et al., 2012). The cooperative of EPOP with Elongin BC to increase gene transcription at actively transcribed genes may additionally involve an EPOP interaction with the deubiquitinase USP7 (Liefke et al., 2016). Analyses of epop knockout (KO) mice, however, revealed no major defects when several of the PRC2-related developmental processes were examined, but epop KO mice exhibited posterior homeotic transformations of the axial skeleton and a shift of the anterior boundary of the expression of certain hox genes (Mocavini et al., 2025). All these observations revealed function for EPOP as a nuclear scaffold protein in early cellular differentiation. Strikingly, we here identify EPOP as a binding partner for the extranuclear, membrane-binding F-BAR protein syndapin I (also called PACSIN1) (Qualmann et al., 1999; Dharmalingam et al. 2009) and reveal a crucial role for the two interacting proteins in dendritic spines of mature neurons. Syndapin I is a neuron-enriched protein that has important roles in both neuromorphogenesis and synaptic functions. In presynapses, syndapin I is crucial for activity-dependent membrane trafficking and shaping of synaptic vesicles (Anggono et al. 2006; Koch et al. 2011). In the postsynaptic compartment, syndapin I was demonstrated to be involved in dendritic spine formation and organization (Schneider et al. 2014) and to regulate the availabilities of both AMPA- and NMDA-type glutamate receptors during basal synaptic activity and during synaptic plasticity, as syndapin I KO altered postsynaptic actin dynamics, synaptic glutamate receptor clustering, as well as the mobility and internalization of AMPA receptors (Koch et al., 2020). As a consequence, syndapin I KO mice displayed schizophrenia-like behavior and epileptic seizures (Koch et al., 2011; Koch et al., 2020). Biochemical, immunocytochemical and immunohistochemical analyses revealed that EPOP, in addition to its nuclear localization, is a somatodendritic protein in mature neurons. Studies in syndapin I KO neurons uncovered that this extranuclear presence of EPOP depended on its direct binding partner, syndapin I. Loss-of-function of EPOP in mature hippocampal neurons negatively impacted the density, morphology and postsynaptic-density composition of mushroom-type dendritic spines. Beyond EPOP’s PRC2-related role in cell differentiation and early development, the observed phenocopy of the postsynaptic phenotypes of syndapin I knockdown in neurons strongly suggest thus far unrecognized, extranuclear functions of EPOP in mature, postmitotic cells, which already are fully integrated into cellular networks. Results EPOP interacts directly with the F-BAR protein syndapin I via an SH3 domain-dependent mechanism Previous studies have identified the F-BAR protein syndapin I as a crucial postsynaptic coordinator in the formation of excitatory synapses and dendritic spines. Syndapin I-enriched membrane nanodomains thereby appear to represent important spatial cues and organizing platforms, shaping dendritic membrane areas into synaptic compartments (Schneider et al., 2014). To deepen the molecular understanding of this crucial role of syndapin I, we performed yeast two-hybrid (Y2H) screens with syndapin I as bait. Interestingly, several clones identified encoded for six independent C-terminal fragments of the mouse protein elongin BC and polycomb repressive complex 2 associated protein (EPOP; NP_780541.2), with the longest Y2H clone #320 representing aa52-369 and the shortest C-terminal fragment (Y2H clone #327) spanning aa222-369 of mouse EPOP (Fig. 1A). Retransformation of the isolated EPOP prey plasmids with syndapin I bait plasmids confirmed the specificity of the Y2H hits by both growth on drop-out plates and β-galactosidase activity. Reporter gene assessments using syndapin I ∆SH3 as bait furthermore revealed that the syndapin I SH3 domain was required for the interaction (Fig. 1B). Reconstitutions with purified proteins demonstrated that a GST-fusion protein of the syndapin I SH3 domain specifically coprecipitated different TrxHis-fusion proteins of EPOP (Fig. 1C and Supplementary Fig. S1 A,B). The syndapin I/EPOP interaction therefore is direct. Coprecipitation analyses with immobilized recombinant GST-syndapin I and deletion mutants thereof together with GFP-EPOP (aa1-369) demonstrated the interaction of the two full-length proteins and furthermore showed that the SH3 domain of syndapin I is both required and sufficient for the association with EPOP (Fig. 1D and Supplementary Fig. S1 C). Coi mmunoprecipitation experiments demonstrated the in vivo-relevance of the EPOP interaction with syndapin I. GFP-syndapin I was coimmunoprecipitated with Flag-tagged EPOP in a specific manner (Fig. 1E). Also in intact cells, EPOP and syndapin I formed protein complexes as demonstrated by specific reconstitution experiments at outer mitochondrial membranes (Fig. 1F-I). In contrast to the recruitment of wild-type syndapin I, mito-mCherry-EPOP-decorated mitochondria did not recruit syndapin I SH3 (Fig. 1I). Thus, also in living, intact cells, a specific association between EPOP and syndapin I can be observed and is critically mediated by the SH3 domain of syndapin I (Fig. 1F-I). The subcellular distribution of EPOP is not restricted to the nucleus but anti-EPOP immunoreactivity can also be observed in the neuronal dendritic arbor To our knowledge, functions described for EPOP have so far been restricted to the nucleus mainly mediated through its association with the polycomb repressive complex 2, which is responsible for histone methylation, and the transcription elongation factor Elongin BC (Bartke et al., 2010; Zhang et al., 2011; Liefke et al., 2016; Beringer et al., 2016). Our identification of syndapin I, a specifically neuron-enriched member of the syndapin subfamily of F-BAR domains proteins, which had only been detected at extranuclear locations within cells, as binding partner of EPOP raised the important question whether the localization of EPOP would be restricted to nuclear compartments or whether EPOP might instead be able to act as a nucleocytoplasmic effector. Biochemical fractionation of mouse brain homogenates analyzed with antibodies raised and affinity-purified against mouse EPOP (aa52-369) (for anti-EPOP antibody characterization see Supplementary Fig. S2 A,B) showed that endogenous EPOP was not only detected in pellet fractions containing cell nuclei, as demonstrated by efficient labeling with antibodies against the nuclear marker laminB1, but also in supernatant fractions devoid of anti-laminB1 signals but instead showing enrichment of the cytoplasmic protein GAPDH (Fig. 2A). Anti-EPOP immunoreactivity was detected as two bands with slightly different apparent molecular weights. EPOP was observed in form of a 50 kD band that was nucleus-enriched and a more abundant 48 kD band, which was mainly detected in the postnuclear supernatant (Fig. 2A). Especially the 50 kD band runs significantly higher than the calculated molecular weight of EPOP (39 kD). Yet, both bands were also visible when overexpressed untagged EPOP was detected by anti-EPOP antibodies. Both bands thus were EPOP-specific, i.e. represent different forms of EPOP originating from the same mRNA (Fig. S2 A). In line, EPOP is not predicted to have splice variants but is encoded by a single exon. Syndapin I was clearly detectable in the supernatant but not in the pellet fraction (Fig. 2A). These experiments thus argued that EPOP/syndapin I complexes are cytosolic and therefore represent a thus far unnoticed function of EPOP that is different from EPOPs known association with the polycomb repressive complex 2 in the nucleus. Syndapin I exhibits high expression levels in various regions of the brain and had been shown to play important roles in hippocampal and cortical neurons (Dharmalingam et al., 2009; Koch et al., 2011; Schneider et al., 2014; Koch et al., 2020). We therefore next analyzed the expression pattern of EPOP in adult brains by both Western blot and immunofluorescence analyses. Immunocytochemistry of sagittal sections of adult mouse brains detected high levels of endogenous EPOP in various regions including the striatum, hippocampus, cortex and cerebellum (Fig. 2B). This rather wide-spread expression of EPOP in the brain was corroborated by Western blot analyses of tissue material from various brain regions. EPOP protein expression in the brain overlapped significantly with the distribution of syndapin I (Fig. 2C). To unequivocally address the localization of endogenous EPOP in neuronal cells, EPOP immunoreactivity was analyzed in comparison to appropriate markers in mouse brain slices. Endogenous EPOP colocalized with MAP2, an exclusively neuronal expressed protein, and not with GFAP, a typical glial marker protein for astrocytes, as demonstrated for the CA1 region of the hippocampus (Fig. 2D). Thus, endogenous EPOP is preferably expressed in neurons. In neurons, EPOP did not only show a nuclear localization and a particularly high presence in nucleoli (Fig. S2 C), but also a cytoplasmic localization. This property of EPOP was observed both in CA1 and the dentate gyrus (Fig. 2E). Strikingly, no colocalization of endogenous EPOP with SMI312, a pan-axonal neurofilament marker, was detectable in both mouse brain sections (Fig. 2E) and in primary hippocampal rat cultures (Fig. 2F). In contrast, EPOP showed clear overlay with the dendritic marker MAP2 (Fig. 2D-F). Together, these observations suggested that EPOP’s extranuclear functions may predominantly be of cytoplasmic and dendritic rather than axonal nature. The EPOP binding partner syndapin I is important for the nucleocytoplasmic distribution of EPOP Our data raised the important question what determines the localization of EPOP to various subcellular compartments of neurons and how and to what extent these distinct EPOP localizations are dependent on the respective binding partners of EPOP. Importantly, the newly identified EPOP binding partner syndapin I influenced EPOP’s nucleocytoplasmic distribution. Ectopic overexpression of syndapin I in COS-7 cells led to elevated levels of GFP-EPOP in the cytoplasm (Fig. 3A). Quantitative analyses showed that the ratio of nuclear to cytoplasmic EPOP dropped by about − 32% when syndapin I was present (Fig. 3A,B). The induced localization change of EPOP from the nuclear to the cytoplasmic compartment by syndapin I presence suggested that syndapin I might act as an extranuclear anchor for EPOP. Coprecipitation experiments indeed demonstrated that endogenous syndapin I was specifically coprecipitated from mouse brain extracts by immobilized EPOP 222–369 (Fig. 3C). Interestingly, this was not the case when an N terminal fragment of EPOP was used suggesting that the endogenous syndapin I protein from brain interacted with the rather proline-rich C terminal half of EPOP (Fig. 3C). We next analyzed whether syndapin I was indeed important for controlling the subcellular localization of endogenous EPOP in neurons. For this purpose, intensity ratios of the immunosignals of EPOP measured in different subcellular compartments in primary hippocampal neurons prepared from wild-type (WT) and syndapin I KO neurons (Koch et al., 2011) were compared (Fig. 3D,E). When the ratios of nuclear EPOP immunoreactivity was related to the whole cell EPOP immunoreactivity, significant nuclear accumulation of EPOP was detected in syndapin I KO neurons compared to WT neurons (Fig. 3F). A significant increase was also observed in the nucleoplasm of syndapin I-deficient neurons (Fig. 3G). In order to further corroborate these findings and to explicitly investigate the cytoplasmic pool of EPOP, we also determined the nuclear anti-EPOP immunolabeling intensities related to the respective extranuclear, i.e. cytoplasmic, pool of EPOP immunoreactivity. Also this ratio showed a significant increase (+ 11%; Fig. 3H). Thus, syndapin I influences the localization of EPOP at the endogenous level. In the presence of syndapin I, EPOP was located relatively less frequently in the nucleus. EPOP is localized to dendritic spines Our studies observed a high expression of EPOP in the adult neuronal tissue and in mature primary cultures. So far, EPOP had been mainly described as an important protein in embryonic stem cells (Zhang et al. 2011, Liefke & Shi 2015). Investigating EPOP protein expression levels in brain homogenates of mice of different age stages, both the 50 kD band corresponding to nuclear-enriched EPOP (red arrow in Fig. 4A; compare Fig. 2A) and corresponding to the 47 kD band of EPOP reflecting the cytoplasmic EPOP (green arrow in Fig. 4A; compare Fig. 2A) could be detected in the E16 embryonic stage. During development, the 50 kD band reflecting nuclear EPOP (red arrow) decreased slightly towards postnatal stage P12. A similar pattern was observed for an anti-EPOP band at 40 kD (black arrow). It may therefore represent a somewhat variable degradation product of nuclear EPOP. In contrast, the 48 kD band of EPOP immunoreactivity (Fig. 4A; green arrow), was found to increase strongly in the postnatal stages. Thus, the form of EPOP corresponding to the cytoplasmic EPOP (compare Fig. 2A) became more abundant during development. Importantly and in line with the identified role of syndapin I as subcellular EPOP distribution modulator (Fig. 3), a similar increase of expression was observed for EPOP’s interaction partner syndapin I (Fig. 4A; Koch et al. 2011). Also in developing cultures of dissociated embryonic neuronal cultures, a comparable increase in expression of particularly the non-nuclear form of EPOP was observable (Fig. 4B). The additional anti-EPOP band at 40 kD showed a developmental decline and was almost undetectable at later postnatal stages (Fig. 4A) and in mature cultured neurons (Fig. 4B). The development-dependent course of expression of EPOP in the neural system showed great similarity to those of the three synaptically occurring proteins syndapin I, ProSAP1 and synapsin 1, which were analyzed in parallel (Fig. 4B). This could reflect some role of EPOP in dendritic spine and/or postsynapse formation. Immunocytochemical studies of endogenous EPOP in hippocampal neurons revealed that EPOP was initially concentrated in the nucleus, but in mature neurons more and more also showed abundance in the MAP2-positive dendrites, in which EPOP overlapped with the localization of syndapin I (Supplementary Fig. S3 ). As an EPOP presence in specifically dendritic spines would allow for a functional cooperation with syndapin I in spine and synapse formation (Schneider et al., 2014), we next addressed whether a portion of the extranuclear EPOP would colocalize with the postsynaptic marker homer1. In addition to the again clear nuclear, somatic and dendritic localization, immunostaining of mature neurons indeed showed colocalizations of endogenous EPOP with homer1 (Fig. 4C). Applying, anti-EPOP and anti-syndapin I antibodies together with antibodies against an additional component of the postsynaptic density (PSD), PSD-95, we furthermore observed that EPOP colocalized with its interaction partner syndapin I in PSD95-positive dendritic spines (Fig. 4D). EPOP is crucial for dendritic spines and for proper spine head organization In order to be able to address whether EPOP may indeed be critical for dendritic spine formation and/or maintenance, we next established EPOP RNAi. Quantitative, fluorescence-based Western blot analyses demonstrated that the expression of EPOP was successfully and significantly suppressed by particularly EPOP RNAi-5 (Fig. 5A). Examinations of mature rat primary neuronal cultures that had been incubated with anisomycin, a protein biosynthesis inhibitor, for different periods of time revealed a rather rapid degradation of EPOP. Quantitative immunoblotting analyses showed that, compared to the DMSO-treated control cultures, a significant reduction in EPOP protein levels (around − 63%) was already detectable after 12 hours of anisomycin treatment. Since almost 75% of EPOP had already been degraded after 24 h (Fig. 5B,C), an incubation time of 48 h was considered as fully sufficient to significantly reduce the endogenous expression of EPOP by RNAi and to analyze the resulting EPOP loss-of-function phenotypes. The depletion of endogenous EPOP by EPOP-RNAi-2 in hippocampal neurons led to a reduction in the density of all dendritic spines by approximately 18% compared to neurons transfected with scrambled RNAi (Fig. 5D,E). This effect was mainly due to a reduction of the density of mushroom-shaped dendritic spines by more than 24% compared to the control (Fig. 5F). In addition, the mushroom-shaped dendritic spines of EPOP-RNAi-2 transfected neurons showed significantly narrower heads (Fig. 5G). Almost the same effects could be observed with EPOP-RNAi-5. In particular the spine head width EPOP loss-of-function phenotype was even more pronounced and thus had a higher significance compared the control scr-RNAi (Fig. 5F,G). The density of all dendritic spines resembled the results of RNAi-2 in absolute numbers relatively well when RNAi-5 was used but showed only a trend towards reduction (Fig. 5E). Analogously to EPOP-RNAi, neurons transfected with syndapin I-RNAi showed somewhat similar dendritic spine phenotypes when compared to EPOP loss-of-function, i.e. a significant reduction in the density of all dendritic spines, especially the mushroom-shaped dendritic spines, and a significant reduction in the head width of mushroom-shaped dendritic spines (Fig. 5D-G). To confirm the specificity of these phenotypes, we performed rescue experiments, in which an RNAi-resistant mutant of EPOP was co-expressed with RNAi-5. Quantitative immunoblot analyses demonstrated that – in contrast to the significant depletion of WT rat EPOP upon coexpression of EPOP-RNAi-5 - coexpressed EPOP* could be detected with a protein level comparable to that under coexpression of scr-RNAi in quantitative anti-EPOP immunoblotting studies (Fig. 6A,B). An IRES system was subsequently used in order to ensure simultaneous expression of the respective RNAi, the reporter mCherryF and, if appropriate, EPOP*. EPOP-RNAi-5_IRES showed similar effects compared to EPOP-RNAi-5 (Fig. 6C-F; compare Fig. 5). The density of all dendritic spines was slightly but not significantly reduced after EPOP-RNAi-5_IRES. A slight increase in the overall dendritic spine density was noted when EPOP-RNAi-5_IRES_EPOP* was expressed, which was not significant compared to the control but when compared to EPOP-RNAi-5_IRES (Fig. 6D). Furthermore, a highly significant reduction in the density of mushroom-type dendritic spines by -24% compared to the control could be observed for RNAi-5. This phenotype was abolished by the expression of EPOP* (Fig. 6E). Also, the reduction in the head width of mushroom-shaped dendritic spines induced by EPOP depletion could be reversed by coexpressing EPOP* and no significant difference from the control were observed anymore (Fig. 6F). Overall, the experiments showed that the dendritic spine phenotypes induced by EPOP depletion with RNAi-5 were rescued by reexpressing EPOP*. This proved the specificity of the identified EPOP loss-of-function phenotypes in dendritic spine formation and organization. Depletion of EPOP causes a reduction of homer1 in dendritic spines The results obtained so far raised the important question whether and to what extent not only the morphology of dendritic spines was affected by the depletion of EPOP, but also its PSD, and thus a possible functional limitation was present. The analysis of the postsynaptic protein homer1 showed a strong impairment in dendritic spines after EPOP RNAi (Fig. 7). EPOP-RNAi-5 resulted in a 48% reduction in homer1 VOI-Rs in dendritic spines compared to control (Fig. 7B). Only 41% of the mushroom-type dendritic spines still showed a homer1 VOI-R after EPOP-RNAi-5, compared to 73% in the control (scr-RNAi; relative: -44%, Fig. 7C). The homer1-positive VOI-Rs that still were detected in dendritic spines furthermore showed a reduced intensity of homer1 and were significantly smaller (Fig. 7E, F). Consequently, depletion of EPOP with RNAi-5 resulted in a drastic reduction of the postsynaptic scaffold component homer1 in dendritic spines. Interestingly, syndapin I RNAi resulted in a similarly clear impairment of the homer1 content in dendritic spines as EPOP-RNAi-5 (Fig. 7). The general occurrence in dendritic spines and explicitly in mushroom-shaped dendritic spines, was massively reduced (Fig. 7B,C). In addition, the intensity of homer1 in the remaining VOI-Rs (Fig. 7D) as well as their volume (Fig. 7E) was significantly reduced. Syndapin I loss-of-function phenotypes in dendritic spine formation and organization thus mirrored those of its interaction partner EPOP in neurons and thereby highlighted their also functional relationship in postsynapses. Both EPOP and syndapin I depletion thus caused a massive reduction in homer1 in mature dendritic spines. Taking into account the reduction of the density of especially mushroom-shaped dendritic spines upon EPOP or syndapin I RNAi (compare Fig. 5), the postsynaptic proteins along the dendrite were even more strongly reduced. Discussion EPOP is a mammalian-specific protein that previously had been defined due to its association with PRC2 (Zhang et al., 2011; De Cegli et al., 2013; Smits et al., 2013; Alekseyenko et al., 2014). Consistent with the PRC2 providing a crucial epigenetic mechanism for regulating embryonic development, EPOPs localization has been reported as nuclear in embryonic stem cells (Liefke & Shi 2015; Behringer et al., 2016; Liefke et al., 2016; Healy et al., 2019). We here report a moonlighting function for EPOP in mature neurons outside the nuclear compartment together with its herein identified binding partner syndapin I. Our data clearly demonstrate that in mature neurons – both in dissociated cultures as well as in intact tissue in various brain regions – EPOP in addition adopts a somatodendritic localization. Biochemical fractionation data confirmed that EPOP is present in both nuclear and extranuclear compartments. The colocalization with MAP2 furthermore demonstrated that EPOP is preferentially expressed in neurons, whereas no prominent overlap with glial cells identified both in cultures and in brain section by anti-GFAP labeling could be observed. It may well be that these observations in mature neurons represent a cell biological specialty, as neurons are both highly differentiated cells and postmitotic, yet they do express relatively high amounts of EPOP. It therefore seems that PRC2 functions in embryonic stem cells and extranuclear functions of the accessory PRC2 component EPOP represent completely distinct functional aspects. The extranuclear role of EPOP in mature neurons appears to involve EPOP interactions with syndapin I. This finding is based on several experimental results. The identification of EPOP/syndapin I complexes by Y2H screening was underscored by coprecipitation of GFP-EPOP from cell lysates by recombinant syndapin I. EPOP associated with the SH3 domain of syndapin I, which was both critical but also sufficient for the interaction. Given the fact that EPOP is rich of prolines and also harbors a variety of PxxP-motifs, it appears likely that the identified interaction is based on a canonical SH3/PxxP association mode. EPOP indeed directly interacted with syndapin I, as proven by in vitro reconstitution with purified components. The in vivo relevance of the interaction was highlighted by heterologous coimmunoprecipitations showing effective association of syndapin I with EPOP. The fact that EPOP was able to coprecipitate endogenous syndapin I from brain extracts furthermore demonstrated the importance of the interaction in the adult brain. Importantly, our mitochondrial recruitment experiments ruled out that the observed syndapin I/EPOP interaction in the biochemical experiments represent putatively possible post-solubilization artifacts due to disruption of nuclear integrity and release of nuclear components. Instead, these experiments visually demonstrated specific complex formation in fully intact cells. The in vivo-recruitment assays also revealed that, if high-affinity extranuclear docking sites for EPOP are provided, the equilibrium of extranuclear and nuclear EPOP can readily be shifted. The mitochondrial recruitment assays are thereby very well in line with our observations of shifts in cytoplasmic/nuclear distribution observed both upon syndapin I overexpression and conversely also upon deficiency of syndapin I in neurons. Several aspects make syndapin I particularly suitable as extranuclear EPOP anchor, it’s efficient, SH3-based association with EPOP, the fact that syndapin I is binding to the plasma membrane via its N-terminal F-BAR domain (Itoh et al., 2005; Dharmalingam et al., 2009) and the special property that syndapin I acts as a dimer (Kessels & Qualmann, 2006), i.e. a syndapin I dimer has the potential to bind to EPOP and to simultaneously associate with further cellular components via an SH3 domain interaction. Remarkably, neither in previous studies nor in comparative biochemical and immunocytochemical analyses in parallel to EPOP, syndapin I has ever been detected in nuclear compartments. It thus seems that syndapin I provides extranuclear anchor points for EPOP and that EPOP undergoes nucleocytoplasmic shuttling. In line, in silico analyses suggest both nuclear localization signals (NLS) and a nuclear export signal (NES) to be present in EPOP (our unpublished data). Comparisons of the immunoreactivity of endogenous EPOP in wildtype (WT) and syndapin I KO neurons revealed that syndapin I is critical for keeping a portion of EPOP extranuclear. Together, our experiments demonstrated that syndapin I is not only capable but – under physiological conditions – also required for proper intracellular distribution of EPOP in mature neurons. Interestingly, EPOP was detected to overlap with the dendritic marker MAP2 but not with the axonal marker SMI312 in both mouse brain sections and in primary hippocampal rat cultures. EPOP thus seems to have a high preference for the dendritic/postsynaptic compartment. This observation distinguishes EPOP from its binding partner syndapin I, which also functions in axonal morphogenesis (Dharmalingam et al., 2009) and in presynapses (Koch et al., 2011). Comparable to its interaction partner syndapin I and further synaptic proteins, such as synapsin I and ProSAP1/Shank2, EPOP expression increases upon neuronal development and showed highest levels in fully developed and differentiated neurons. Our observations at the protein level are in line with in situ hybridizations in mouse brains, that also highlighted a prominent presence of EPOP mRNA in cortical and hippocampal neurons not only at the embryonic but also postnatally, at a juvenile age (P21) (De Cegli et al. 2013). Upon in vitro differentiation of mouse embryonic stem cells, in contrast, a decline of EPOP was observed (De Cegli et al., 2013; Liefke et al., 2016). This apparent difference to our result in differentiating neurons supports a distinct role of EPOP in differentiated neurons independent form its earlier reported functions in PRC2 and ElonginB/C regulation in the nucleus. In line with such a distinct role, at least in neurons, the main proteins of the canonical PRC2, such as enhancer of zeste homolog (EZH) 2 or suppressor of zeste 12, exhibited a rapid decrease in expression during neuronal development (Henriquez et al., 2013). EPOP has been precipitated with both EZH2 and EZH1, although EZH1 and EZH2 can be excluded as simultaneous components of a PRC2 (Zhang et al., 2011; Alekseyenko et al., 2014; Margueron et al., 2008; Shen et al., 2008). Interestingly, EZH2 induces the expression of PSD95, a postsynaptic scaffold protein (Henriquez et al., 2013). Yet, only for EZH1, a non-canonical major protein of PRC2, a more long-lasting neuronal expression has been described (Henriquez et al., 2013). EPOP as well as its interaction partner syndapin I were observed to rise in expression towards neuronal maturation and towards the formation of neuronal networks by synapse formations. In the dendritic compartment of mature neurons, EPOP was observed to localize to dendritic spines, where its immunoreactivity overlapped with the PSD proteins PSD95 and homer1 and with its binding partner syndapin I. The dendritic spine offers several specific aspects bringing about elevated levels of physically attached syndapin I. Syndapin I interacts with the postsynaptic scaffold protein ProSAP1/Shank2 (Schneider et al., 2014) and syndapin I’s membrane anchoring shows a preference for dendritic spines and their different membrane curvatures (Schneider et al., 2014). Loss-of-function studies in primary hippocampal neurons demonstrated that EPOP was particularly important for mushroom-type spines. RNAi-mediated knockdown of EPOP significantly reduced the density of mushroom spines, a phenotype consistently observed for two different RNAi tools. Rescue experiments with an RNAi-resistant version of EPOP showed the specificity of the loss-of-function analyses. Interestingly, also knockdown of the EPOP interaction partner syndapin I resulted in a decreased mushroom spine density in DIV18 neurons. In previous studies, similar defects had been observed in less mature dissociated hippocampal neurons (DIV14) (Schneider et al., 2014). EPOP loss-of-function – and consistently likewise syndapin I loss-of-function – did not only result in fewer mushroom spines, but those present had significantly decreased spine head sizes. The dendritic spine heads harbor the postsynaptic scaffold and signal integration and organization machinery. Impairments in PSD organization upon syndapin I-deficiency had been described previously. Syndapin I deficient neurons showed a significant reduction of PSD95 points along the dendrites of younger, DIV14 neurons (Schneider et al. 2014). Since previous studies had reported influences of PRC2 components on PSD95 mRNA expression levels through transcriptional control (Henriquez et al., 2013), we did not address putative phenocopying defects upon EPOP deficiency by analyzing PSD-95. Instead, we analyzed the major synaptic scaffold protein homer1 as an indicator for proper PSD organization. Analyses of homer1 and the morphology of spines clearly showed that the depletion of endogenous EPOP, analogous to a depletion of syndapin I, resulted in a massive impairment of mature spines and their postsynaptic organization. The mushroom spines still remaining upon EPOP RNAi or syndapin I RNAi exhibited additional defects in the composition and organization of the postsynaptic density, they contained significantly lesser or even no detectable levels of the major synaptic scaffold protein homer1. Both the volume and the mean intensity of homer1-positive structures in mushroom spine heads were highly decreased. The fact that upon loss of EPOP or syndapin I, the integrity of dendritic spines and of postsynapses was severely impaired indicated that both proteins not only play an important role in the formation of spines but are also responsible for the maintenance and stabilization of mature spines. The broad correspondence of the phenotypes on spines and postsynapses induced by syndapin I and EPOP RNAi and their identified direct interaction strongly suggests that EPOP and syndapin I jointly stabilize mature spines as well as their postsynapses. Material and methods Plasmids Full-length mouse ( mus musculus ) EPOP (NCBI NP_780541.1) constructs were generated by performing PCR reactions on a mouse cDNA library ( BD Biosciences λTriplEx™) and subsequently cloned into pCMV-Tag2B (Stratagene), pEGFP-C2 (Clontech) and pIRES2-eGFP (BD Bioscience/Clontech). For mitochondrially targeted EPOP constructs, EPOP was subcloned into Mito-Flag-mCherry-pCMV-Tag2B (Mito-mCherry), a mitochondrial targeting vector described previously (Hou et al., 2015). Further EPOP constructs were obtained by subcloning inserts from Y2H clones into pGEX-4T2, pEGFP-CW3 and pET32c (Novagen). GFP-EPOP 1–169 was amplified by PCR using mouse full-length EPOP as template with the primers 5’-cggaattcatggagactctgtgtcctcct-3’ and 5’-cgcgtcgacctaactgctagctgcatcaagacc-3’ and subcloned into pEGFP-C2. A corresponding GST-fusion was obtained by subcloning into pGEX-5X-1. Rat ( rattus norvegicus ) EPOP (NCBI NP_001103097.1 with G117V und R199P sequence variants corresponding to the sequence variants deposited for M0RD31_RAT in UniProt variant viewer) was amplified from adult rat cortex cDNA with the primers 5’-ccggaattcatggagactctgtgtcc-3’ and 5‘-acgcgtcgactcagagttcttccaag-3‘ and subcloned into pEGFP-C2 and pIRES2-eGFP (BD Bioscience/Clontech). RNAi constructs directed against EPOP were generated according to the methods described previously (Ahuja et al., 2007). In brief, phosphorylated primers for EPOP-RNAi-2 (target sequence rat nt 761–785), 5’- gatccaaacttggagtgtccagggcgaaccttgatatccgggttcgccctggacactccaagttttttttta-3’ and 5’- agcttaaaaaaaaacttggagtgtccagggcgaacccggatatcaaggttcgccctggacactccaagtttg-3‘ and EPOP-RNAi-5 (target sequence rat nt 741–761) 5‘-GATCCACTTTGGTGTTACGCGAAAGGACTCGAGAcctttcgcgtaacaccaaagtTTTTTA-3’ and 5’- AGCTTAAAAAACTTTGGTGTTACGCGAAAGGTCTCGAGTcctttcgcgtaacaccaaagtG-3’ were annealed. Subsequently, the products were subcloned into pRNAT-H1.1 coexpressing farnesylated mCherry (mCherry-F) (pRNAT-H1.1/mCherryF) or eGFP (pRNAT-H1.1/GFP). Corresponding RNAi vectors expressing a scrambled RNAi sequence (scr-RNAi) served as controls (Nolze et al., 2013). For an analysis of knockdown efficiency of coexpressed rat EPOP in HEK293 cells via Western blot analyses, rat EPOP was subcloned from pIRES2-eGFP together with the IRES sequence using AfeI/NotI into pRNAT-H1.1/GFP expressing scr-RNAi, EPOP-RNAi-2 and EPOP-RNAi-5, respectively, generating scr-RNAi_GFP_IRES_EPOP, EPOP-RNAi-2_GFP_IRES_EPOP, and EPOP-RNAi-5_GFP_IRES_EPOP, respectively. For rescue experiments, the IRES sequence from pIRES2-eGFP was amplified using the primers 5‘-atagagctcgatccgcccctctccctccccccc-3‘ and 5‘-atacccgggttaattaattaactagttgtggccatattatcatcgtgttt-3‘ and cloned into scr-RNAi and EPOP-RNAi-5 in pRNAT-H1.1/mCherryF, respectively, generating scr-RNAi_IRES and EPOP-RNAi-5_IRES, respectively. RNAi-resistant mouse EPOP containing silent mutations (EPOP*) was generated by mutagenesis PCR using the mutagenesis primers 5’-aaccgcctgatccgccgaagtaagttatggtgctatgccaagggcttcgccctggacact-3’ and 5’-agtgtccagggcgaagcccttggcatagcaccataacttacttcggcggatcaggcggtt-3’ and subcloned into pIRES2-eGFP and subsequently together with the IRES sequence using AfeI/NotI into pRNAT-H1.1/GFP expressing scr-RNAi and EPOP-RNAi-5, respectively, generating scr-RNAi_GFP_IRES-EPOP* and EPOP-RNAi-5_GFP_IRES-EPOP* respectively, and into EPOP-RNAi-5_IRES generating EPOP-RNAi-5_IRES_EPOP*. BD-syndapin I and BD-syndapin I ΔSH3 cloned in the pGBTK7 vector have been described in Braun et al. (2005). GST-syndapin (Izadi et al., 2021), GST-syndapin I SH3 (Braun et al., 2005) and GST-syndapin I ΔSH3 (Qualmann et al., 1999) were described previously. GST-syndapin I BAR (aa1-305) was amplified by PCR ad subcloned into pGEX-5X-1. GFP-syndapin I was described in Kessels & Qualmann (2006). Xpress-tagged syndapin I and syndapin I ΔSH3 were described in Qualmann & Kelly (2000). RNAi constructs against syndapin I have been described previously (Dharmalingam et al., 2009). All PCR-based constructs were verified by sequencing. Proteins GST- and TrxHis-tagged fusion proteins were purified from E. coli as described previously (Qualmann & Kelly, 2000; Schwintzer et al., 2011). Antibodies Antisera against EPOP were raised in rabbit (Pineda Antikörper-Service, Germany) using a purified GST-fusion protein of amino acids 52–369 of mouse EPOP as antigen. Using recombinant TrxHis-EPOP 52–369 immobilized on a CNBr sepharose 4b (Amersham) column and acidic elution, affinity-purified polyclonal anti-EPOP antibodies were obtained. Polyclonal rabbit and guinea pig antibodies against GST-syndapin I as well as anti-GST and anti-TrxHis antibodies were purified from antisera as described previously (Qualmann et al., 1999; Braun et al., 2005; Izadi et al., 2021). Monoclonal mouse anti-GFP antibodies (JL-8) were from Clontech/Takara (632381, Takara). Monoclonal anti-Xpress antibodies (R910-25) were from Invitrogen. Polyclonal rabbit anti-GFP antibodies (ab290) were from Abcam. Monoclonal mouse (M2; F3165) and polyclonal rabbit anti-Flag (F7425) antibodies, monoclonal mouse anti-MAP2 (HM-2; M4403), monoclonal mouse anti-ß-tubulin (clone TUB 2.1) anti monoclonal mouse anti-GFAP (G3893) antibodies as well as polyclonal rabbit anti-ProSAP1 (anti-SHNAK2; HPA008174) antibodies were from Sigma-Aldrich® Co. LLC. Guinea pig anti-MAP2 and anti-homer1 antisera and mouse monoclonal anti-synapsin1 antibodies (clone 46.1) from Synaptic Systems. Mouse monoclonal anti-fibrillarin antibodies (clone 38F3). Mouse monoclonal anti-PSD95 (clone 6G6-1C9) antibodies and rabbit polyclonal anti-laminB1 antibodies were from Abcam. Goat polyclonal anti-GAPDH antibodies were from Santa Cruz Biotechnology. Mouse monoclonal anti-SMI312 (837904) antibodies were from Covance/Biozol. Normal rabbit IgG (10500C) was from Thermo Fischer Scientific. Secondary antibodies included Alexa Fluor488-labeled goat anti-mouse (A-10680), donkey anti-mouse (A-21202), Alexa Flour586-labeled donkey anti-mouse (A10037), donkey anti-rabbit (A10042), Alexa Flour647-labeled donkey anti-mouse (A-31571) and donkey anti-rabbit (A-31573) antibodies (Thermo Fisher Scientific) and Alexa Flour647-labeled donkey anti-guinea pig antibodies (06-605-148, Dianova). Further secondary antibodies used included AlexaFluor680-labeled goat anti-rabbit and anti-mouse antibodies and AlexaFluor680-labeled donkey anti-goat antibodies (A-32734, A-21058; A-21084; Thermo Fisher Scientific); DyLight800-conjugated goat anti-rabbit and anti-mouse antibodies (SA5-35571 and SA5-35521; Thermo Fisher Scientific), donkey anti-guinea pig antibodies coupled to IRDye800, (926-32411; LI-COR Bioscience) and peroxidase-conjugated goat anti-rabbit antibodies (111-035-045; Dianova). Isolation of RNA and reverse transcription The RNA isolation from adult rat cortex was carried out according to Haag et al. (2012). For reverse transcription the RevertAid H minus First Strand cDNA Synthesis Kit was used according to the manufacturer's instructions. In vitro reconstitutions of direct protein–protein interactions Direct protein–protein interactions were demonstrated by coprecipitations with combinations of recombinant TrxHis- and GST-tagged fusion proteins purified from E. coli as described in Izadi et al. (2018). In brief, complex formation of TrxHis-EPOP 52–369 and TrxHis-EPOP 222–369 with GST-syndapin I SH3 was demonstrated in 10 mM HEPES, pH 7.4, 300 mM NaCl, 0.1 mM MgCl 2 , and 1% (v/v) Triton X-100 supplemented with EDTA-free protease inhibitor cocktail. Eluted proteins were analyzed by SDS-PAGE, transferred to polyvinylidene fluoride membranes and then analyzed by immunodetection with anti-TrxHis and anti-GST antibodies by using a Licor Odyssey System. Preparation of brain extracts Rat brains were homogenized with an ultraturrax in HEPES buffer (10 mM HEPES, 1 mM EGTA, 0.1 mM MgCl 2 , 0.15 M NaCl, 1x protease inhibitor without EDTA, pH 7.5; 3 ml per g tissue) for several times for 15–20 seconds at 20000 rpm under ice cooling. The supernatant of a subsequent centrifugation at 150000 x g and 4°C for 45 minutes was obtained as rat brain extract. Preparation and fractionation of mouse tissue homogenates Tissue homogenates in RIPA buffer were prepared as described previously (Hou et al., 2018). For fractionation of mouse cortex homogenates, mouse cortices were prepared and homogenized by 12 strokes with the Potter S homogenizer at 900 min − 1 in 10 µl homogenization buffer (5 mM HEPES pH7.4, 1 x PIC EDTA-free, 0.32 M sucrose, 1 mM EDTA) per mg wet weight. The homogenate was centrifuged for 10 min at 1000 x g and 4°C. Then, 100 µl as supernatant were removed. The sediment was homogenized in the same volume as before. The mixture was then centrifuged for 10 min at 1000 x g and 4°C. The pellet was resuspended in 1 ml PBS / PIC. The fractions were analyzed by SDS-PAGE and Western blot. Lysis of primary cortical rat neurons Cultured neurons were washed once with cold PBS, detached using a cell scraper and taken up in 1 ml cold PBS. The cells were centrifuged for 5 min at 1000 x g and 4° C and the pellet was resuspended in 400 µl lysis buffer (10 mM HEPES, 1 mM EGTA, 0,1 mM MgCl 2 , 10 mM NaCl, 1% (v/v) Triton® X-100, 1 x PIC complete; pH 7.4). After 5 ultrasound pulses (1 s each) and 20 min incubation at 4 ° C on a rotary wheel, the samples were centrifuged at 4°C for 10 min at 20800 x g. Subsequently, 400 µl of the supernatant were mixed with 1.6 ml -20°C cold acetone. The sample was incubated overnight at -20°C. The precipitated proteins were sedimented (30 min at 12000 x g and 4°C) and the pellet was washed with 2 ml of -20° C cold 80% (v/v) ethanol. After drying for two minutes, the pellet was resuspended in 100 µl 1 x SDS sample buffer. In order to facilitate the resuspension, 3 ultrasound pulses of 1 s were applied. Finally, the samples were incubated for 5 min at 95° C and subsequently analyzed by immunoblotting. For the analysis of the time course of degradation of endogenous EPOP, cortical neurons (DIV17) were treated for different time intervals (2–24 h) with the protein synthesis inhibitor anisomycin (7.5 µM final). Cell lysates were prepared as described above. Lysates of neuronal cultures incubated with DMSO (0.2% (v/v) final) were used as solvent control. Yeast two-hybrid analyses Yeast two-hybrid screening was carried out using the GAL4-based Matchmaker yeast two-hybrid system 3 (BD Biosciences Clontech) with full-length syndapin I as a bait and a pretransformed mouse brain library (BD Biosciences Clontech) as described in Braun et al. (2005). Subsequently, prey plasmids were isolated, retransformed into yeast, and mated with yeast strains transformed with BD-syndapin I, BD-syndapin I ΔSH3 , and with the pGBTK7 vector encoding for the BD domain alone, respectively. The obtained diploids were subsequently assayed for the activation of reporter genes according to Kessels & Qualmann (2002). Animals Rats (Crl:WI; Charles River) were used for primary cell culture preparations and biochemical analyses from brain tissue. The generation and analysis of syndapin I KO mice has been described previously (Koch et al., 2011; Koch et al., 2020). All animal breeding was in strict compliance with the EU directives 86/609/EWG and 2007/526/EG guidelines for animal experiments and all related procedures were approved by the local government (Thüringer Landesamt, Bad Langensalza, Germany, and Landesverwaltungsamt, breeding allowance 02-0571/14 and UKJ-17-021). Cell culture, transfection and immunostaining HEK293 and COS-7 cells were maintained in 10 ml DMEM containing 2 mM L-glutamine, 10% (v/v) fetal bovine serum and penicillin/streptomycin (Invitrogen) at 37°C, 90% humidity and 5% CO 2 . Cells for immunofluorescence-based analyzes were processed on coverslips coated with poly-D-lysine. HEK293 and COS-7 cells were transfected with plasmid DNA using TurboFect (Thermo Fisher Scientific) according to manufacturer´s instructions. Immunolabeling of cultured cells was essentially as described in Hou et al. (2018). Mitochondria of living cells were visualized by incubating the cells with a final concentration of 0.2 µM MitoTracker Deep Red 633 for 1 h prior to fixation. Preparation of HEK293 cell lysates for coprecipitation and coimmunoprecipitation analyses was performed as described in Qualmann & Kessels (2006), Schwintzer et al. (2011) and Izadi et al. (2018). Coprecipitation analyses Coprecipitation analyses were performed with HEK293 cell lysates containing overexpressed GFP-fusion proteins or with mouse brain extracts and GST-tagged proteins immobilized to glutathione resin (Genscript Corp.) as described before (Qualmann et al., 1999; Schwintzer et al., 2011). Coimmunoprecipitation Coi mmunoprecipitation analyses of Flag-EPOP and GFP-syndapin I were performed from lysates from transfected HEK293 cells as described previously in Schwintzer et al. (2011). A lysis buffer containing 120 mM NaCl was used. Preparation and cultivation of primary neuronal cultures Preparation of primary rat neurons and their cultivation was carried out according to the protocols of Banker and Cowan (1977) and Brewer et al. (1993). The protocol used was as described in Schneider et al. (2014). Primary hippocampal mouse neurons (DIV17) of wild-type and syndapin I-deficient animals were prepared as described in Koch et al. (2020). Rat cortical neurons were prepared and cultured as described in Wolf et al. (2019). Immunolabeling of cultured cells was essentially as described in Schwintzer et al. (2011) and Schneider et al. (2014). Transient transfection of rat neurons After 16 days in culture, calcium phosphate transfections were performed as described previously (Schneider et al., 2014). For some experiments, the DIV16 neurons were alternatively transiently transfected using Lipofectamine® 2000 as described before (Schwintzer et al., 2011; Qualmann et al., 2004). Immunolabeling of mouse brain sections Mouse brain sections were prepared essentially as described in Haag et al. (2012) and Schneider et al. (2014). Immunolabeling of brain sections was performed according to Schneider et al. (2014) except that PBS instead of PB was used for all solutions. Light microscopy The images of cells and brain sections were taken using a Zeiss AxioObserver.Z1 microscope. In order to achieve pseudo confocality, Z stacks were recorded with an ApoTome, at distances of 0.24 µm (63x/1.4 objective), 0.3 µm (40x/1.3 objective) and 1 µm (20x/0.5 objective), respectively. Images were recorded digitally and either AxioVision or Zen software was used for image processing. For the overview picture of a mouse brain, the Tiles mode of the Zen software was applied with an overlap of 10%; all tiles in the 3 Z planes were recorded at a distance of 10 µm. All single images were stitched together by automatic stitching using Zen software. The overview is shown as maximum intensity projection (MIP). Investigation of the subcellular distribution of overexpressed EPOP in COS-7 cells COS-7 cells were cotransfected with a plasmid encoding for GFP-EPOP in combination with mCherryF or Xpress-syndapin I. Anti-EPOP intensity measurements were based on anti-EPOP immunostaining using Fiji. As a marker for nuclear nuclei acids, DAPI was applied. The entire cell was marked by the reporters mCherryF or syndapin I, respectively. The mean intensity for the cytoplasm was calculated from the mean intensities and the areas of both regions (whole cell, nucleus) by subtraction. Distribution analysis of endogenous EPOP in wild-type and syndapin I-deficient neurons Primary hippocampal mouse neurons (DIV17) of wild-type and syndapin I-deficient animals prepared as described in Koch et al. (2020) were fixed and stained for EPOP, MAP2 and fibrillarin as well as with DAPI and subjected to pseudoconfocal imaging (ApoTome). The evaluations of the Z stacks were carried out blindly. Using Imaris software, fibrillarin and MAP2 immunoreactivity as well as DAPI signals were used to create so-called surfaces of volume-of-interest reconstructions (VOI-Rs). These surface reconstructions represented either the nucleoli (fibrillarin), the nucleus (DAPI) or the (somatodendritic) cytoplasm (MAP2). The mean intensity of endogenous anti-EPOP immunosignals was then measured in these VOI-Rs. Taking into account the VOI-R volume, the mean intensities of the nucleoplasm and of the entire cells were calculated. Subsequently, the ratio of the mean intensities of two compartments was calculated. Morphological analysis of spines Spines were analyzed using rat hippocampal neurons that were transfected at DIV16 and further cultured for 48 h. The neurons were then immunostained and imaged as described above. One dendrite section having a minimum length of 100 µm was evaluated per cell. The investigations were carried out blindly. Imaris software was used to reconstruct and define all spines of the dendritic segment. The diameter of both the dendrite and the spine was calculated using the shortest distance algorithm. The parameters for the detection of the spines were selected as follows: minimum diameter 0.1 µm and maximum length 5 µm. If necessary, manual corrections were made to trace individual spines correctly in 3D. The spines were then grouped using the spine classifier according to their subtypes. The following settings were made for this: Base of the spines: proximal 25% of the spines; head of the spine: distal 25% of the spine; stubby spines: spines length ≤ 0.75 µm; mushroom-shaped spines: spines length > 0.75 µm and maximum head width > 0.5 µm; thin spines: spines length > 0.75 µm and maximum head width ≤ 0.5 µm and maximum head width > 0.3 µm and average neck width < maximum head width; filopodia-like spines: all other spines detected. All parameters were normalized to the values for the respective control in the blinded analyses. Investigation of postsynaptic proteins in spines To investigate the expression of homer1 in the spines, a created Imaris reconstruction ( filament ) as well as only the dendrite of the filament were each converted into individual fluorescence channels. A VOI-R of the dendrite was then created. This was used for masking (voxel within the VOI-R to 0) of the filament channel, so that a new fluorescence channel, which contained only spines, could be generated. A colocalization channel was then formed from the spines channel and the homer1 channel. On the basis of this colocalization channel, VOI-Rs for homer1 were generated, which reflected the signal on postsynaptically localized protein above a manually defined, constant threshold value. Statistics All statistical evaluations carried out in this work were carried out in GraphPad Prism. Whenever applicable (n number restriction), the Shapiro-Wilk test was used to first check the values for normal distribution. Otherwise, the Kolmogorov–Smirnov normality distribution test was used. If the values were distributed normally, a t-test (2 conditions to compare) and a one-way ANOVA (≥ 3 conditions to compare), respectively, were used. For not normally distributed data, a Mann-Whitney U test (2 samples) or a Kruskal-Wallis test (≥ 3 samples) was carried out. Both multiple comparisons were followed by posttests, Bonferroni's Multiple Comparison Test for one-way ANOVA evaluations and a Dunn's Multiple Comparison Test for Kruskal-Wallis tests. Two-way ANOVA examinations were carried out for variance analyses with two factors. Declarations Acknowledgments We thank B. Schade A. Kreusch, M. Öhler and C. Scharf for excellent technical assistance and D. Koch, E. Wünsche and A. Inciute for experimental support. This work was supported by DFG grants QU116/5-2 and 11-1 to BQ and KE685/7-1 to MMK. Author Contributions J.D. conducted the vast majority of experiments shown. R.D. made initial observations and also conducted experiments. N.K. generated primary mouse cultures.J.D, M.M.K, and B.Q. interpreted and visualized data. J.D. cowrote an initial draft of the manuscript. B.Q. and M.M.K conceived the project, provided scientific supervision and funding and wrote the manuscript. Declaration of interests: The authors declare no competing financial interests. References Aguilar, R., Bustos, F. J., Saez, M., Rojas, A., Allende, M. L., van Wijnen, A. J., van Zundert, B. & Montecino, M. Polycomb PRC2 complex mediates epigenetic silencing of a critical osteogenic master regulator in the hippocampus. Biochim. Biophys. Acta 1859 , 1043–1055 (2016). Ahuja, R., Pinyol, R., Reichenbach, N., Custer, L., Klingensmith, J., Kessels, M.M. & Qualmann, B. Cordon-bleu is an actin nucleation factor and controls neuronal morphology. Cell 131 , 337–350 (2007). Alekseyenko, A. A., Gorchakov, A. A., Kharchenko, P. V. & Kuroda, M. I. Reciprocal interactions of human C10orf12 and C17orf96 with PRC2 revealed by BioTAP-XL cross-linking and affinity purification . Proc. Natl. Acad. Sci. USA 111 , 2488–2493 (2014). Anggono, V., Smillie, K. J., Graham, M. E., Valova, V. A., Cousin, M. A. & Robinson, P. J. Syndapin I is the phosphorylation-regulated dynamin I partner in synaptic vesicle endocytosis. Nat. Neurosci. 9 , 752–760 (2006). Banker, G. A. & Cowan, W. M. Rat hippocampal neurons in dispersed cell culture. Brain Res. 126 (1977). Bartke, T., Vermeulen, M., Xhemalce, B., Robson, S. C., Mann, M. & Kouzarides, T. Nucleosome-interacting proteins regulated by DNA and histone methylation. Cell 143 , 470–484 (2010). Beringer, M., Pisano, P., Di Carlo, V., Blanco, E., Chammas, P., Vizán, P., Gutiérrez, A., Aranda, S., Payer, B., Wierer, M. & Di Croce, L. EPOP Functionally Links Elongin and Polycomb in Pluripotent Stem Cells. Mol. Cell 64 , 645–658 (2016). Braun, A., Pinyol, R., Dahlhaus, R., Koch, D., Fonarev, P., Grant, B. D., Kessels, M. M. & Qualmann, B. EHD proteins associate with syndapin I and II and such interactions play a crucial role in endosomal recycling. Mol. Biol. Cell 16 , 3642-3658 (2005). Brewer, G. J., Torricelli, J. R., Evege, E. K. & Price, P. J. Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination. J. Neurosci. Res . 35 , 567-576 (1993). De Cegli, R., Iacobacci, S., Flore, G., Gambardella, G., Mao, L., Cutillo, L., Lauria, M., Klose, J., Illingworth, E., Banfi, S. & di Bernardo, D. Reverse engineering a mouse embryonic stem cell-specific transcriptional network reveals a new modulator of neuronal differentiation. Nucleic Acids Res. 41 , 711–726 (2013). Dharmalingam, E., Haeckel, A., Pinyol, R., Schwintzer, L., Koch, D., Kessels, M.M. & Qualmann, B. F-BAR proteins of the syndapin family shape the plasma membrane and are crucial for neuromorphogenesis. J. Neurosci. 29 , 13315–13327 (2009). Fischer, S., Weber, L. M. & Liefke, R. Evolutionary adaptation of the Polycomb repressive complex 2. Epigenetics Chromatin . 15, 7 (2022). Gong, L., Liu, X., Yang, X., Yu, Z., Bhat, M. Y., Chen, S., Lemoff, A., Xing, C. & Liu, X. EPOP restricts PRC2.1 targeting to chromatin by directly modulating enzyme complex dimerization. Nat. Commun. 17, 1565 (2026). Granat, J., Liu, S., Popoca, L., Oksuz, O. & Reinberg, D. EPOP and MTF2 activate PRC2 activity through DNA-sequence specificity . Proc. Natl. Acad. Sci. USA 123 , e2527303123 (2026). Haag, N., Schwintzer, L., Ahuja, R., Koch, N., Grimm, J., Heuer, H., Qualmann, B. & Kessels, M. M. The actin nucleator Cobl is crucial for Purkinje cell development and works in close conjunction with the F-actin binding protein Abp1. J. Neurosci. 32 , 17842–17856 (2012). Healy, E., Mucha, M., Glancy, E., Fitzpatrick, D. J., Conway, E., Neikes, H. K., Monger, C., Van Mierlo, G., Baltissen, M. P., Koseki, Y., Vermeulen, M., Koseki, H. & Bracken, A. P.. PRC2.1 and PRC2.2 Synergize to Coordinate H3K27 Trimethylation. Mol. Cell 76 , 437–452 (2019). Henriquez, B., Bustos, F. J., Aguilar, R., Becerra, A., Simon, F., Montecino, M. & van Zundert, B. Ezh1 and Ezh2 differentially regulate PSD-95 gene transcription in developing hippocampal neurons. Mol. Cell. Neurosci. 57 , 130–143 (2013). Hou, W., Izadi, M., Nemitz, S., Haag, N., Kessels, M.M. & Qualmann, B. The actin nucleator Cobl is controlled by calcium and calmodulin. PLoS Biol. 13 , e1002233 (2015). Hou, W., Nemitz, S., Schopper, S., Nielsen, M.L., Kessels, M.M. & Qualmann, B. Arginine methylation by PRMT2 controls the functions of the actin nucleator Cobl. Dev. Cell 45 , 262–275 (2018). Itoh, T., Erdmann, K. S., Roux, A., Habermann, B., Werner, H. & De Camilli, P. Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev. Cell 9 , 791–804 (2005). Izadi, M., Schlobinski, D., Lahr, M., Schwintzer, L., Qualmann, B. & Kessels, M.M. Cobl-like promotes actin filament formation and dendritic branching using only a single WH2 domain. J. Cell Biol. 217 , 211-230 (2018). Izadi, M., Seemann, E., Schlobinski, D., Schwintzer, L., Qualmann, B. & Kessels, M.M. Functional interdependence of the actin nucleator Cobl and Cobl-like in dendritic arbor development. eLife 10 , e67718 (2021). Kessels, M.M. & Qualmann, B. Syndapins integrate N-WASP in receptor-mediated endocytosis. EMBO J. 21 , 6083–6094 (2002). Kessels, M.M. & Qualmann, B. Syndapin oligomers interconnect the machineries for endocytic vesicle formation and actin polymerization. J. Biol. Chem. 281 , 13285-13299 (2006). Koch, N., Koch, D., Krueger, S., Tröger, J., Sabanov, V., Ahmed, T., McMillan, L.E., Wolf, D., Montag, D., Kessels, M.M., Balschun, D. & Qualmann, B. Syndapin I loss-of-function in mice leads to schizophrenia-like symptoms. Cereb. Cortex 30 , 4306–4324 (2020). Koch, D., Spiwoks-Becker, I., Sabanov, V., Sinning, A., Dugladze, T., Stellmacher, A., Ahuja, R., Grimm, J., Schüler, S., Müller, A., Angenstein, F., Ahmed, T., Diesler, A., Moser, M., tom Dieck, S., Spessert, R., Boeckers, T.M., Fässler, R., Hübner, C.A., Balschun, D., Gloveli, T., Kessels, M.M. & Qualmann, B. Proper synaptic vesicle formation and neuronal network activity critically rely on syndapin I. EMBO J. 30 , 4955-4969 (2011). Liefke, R. & Shi, Y. The PRC2-associated factor C17orf96 is a novel CpG island regulator in mouse ES cells. Cell Discov. 1 , 15008 (2015). Liefke, R., Karwacki-Neisius, V. & Shi, Y. EPOP Interacts with Elongin BC and USP7 to Modulate the Chromatin Landscape. Mol. Cell 64 , 659–672 (2016). Liu, P. P., Xu, Y. J., Teng, Z. Q. & Liu, C. M. Polycomb Repressive Complex 2: Emerging Roles in the Central Nervous System. Neuroscientist 24 , 208–220 (2018). Mocavini, I., Mallol, A., Gutierrez, A., Chammas, P., Carretero, A., Dias, A., Blanco, E., Rodriguez Arce, I., Mallo, M., Di Croce, L. & Payer, B. The PRC2-associated factor EPOP is required for Hox gene regulation during axial development in mice. Dev. Biol. 527 , 250–259 (2025). Nolze, A., Schneider, J., Keil, R., Lederer, M., Hüttelmaier, S., Kessels, M. M., Qualmann, B. & Hatzfeld, M. FMRP regulates actin filament organization via the armadillo protein p0071. RNA 19 , 1483–1496 (2013). Okumura, F., Matsuzaki, M., Nakatsukasa, K. & Kamura, T. The Role of Elongin BC-Containing Ubiquitin Ligases. Front. Oncol. 2 , 10 (2012). Qualmann, B., Roos, J., DiGregorio, P.J. & Kelly, R.B. Syndapin I, a synaptic dynamin-binding protein that associates with the neural Wiskott-Aldrich syndrome protein. Mol. Biol. Cell 10, 501–513 (1999). Qualmann, B., Boeckers, T. M., Jeromin, M., Gundelfinger, E. D. & Kessels, M. M. Linkage of the actin cytoskeleton to the postsynaptic density via direct interactions of Abp1 with the ProSAP/Shank family . J. Neurosci. 24 , 2481-2495 (2004). Qualmann, B. & Kelly, R.B. Syndapin isoforms participate in receptor-mediated endocytosis and actin organization. J. Cell Biol. 148, 1047–1062 (2000). Schneider, K., Seemann, E., Liebmann, L., Ahuja, R., Koch, D., Westermann, M., Hübner, C.A., Kessels, M.M., and Qualmann, B. (2014). ProSAP1 and membrane nanodomain-associated syndapin I promote postsynapse formation and function. J. Cell Biol. 205 , 197-215 (2014). Schwintzer, L., Koch, N., Ahuja, R., Grimm, J., Kessels, M.M. & Qualmann, B. The functions of the actin nucleator Cobl in cellular morphogenesis critically depend on syndapin I . EMBO J. 30 , 3147–3159 (2011). Smits, A. H., Jansen, P. W., Poser, I., Hyman, A. A. & Vermeulen, M. Stoichiometry of chromatin-associated protein complexes revealed by label-free quantitative mass spectrometry-based proteomics. Nucleic Acids Res. 41 , e28 (2013). Wolf, D., Hofbrucker-MacKenzie, S.A., Izadi, M., Seemann, E., Steiniger, F., Schwintzer, L., Koch, D., Kessels, M.M. & Qualmann, B. Ankyrin repeat-containing N-Ank proteins shape cellular membranes. Nat. Cell Biol. 21 , 1191-1205 (2019). Zhang, Z., Jones, A., Sun, C. W., Li, C., Chang, C. W., Joo, H. Y., Dai, Q., Mysliwiec, M. R., Wu, L. C., Guo, Y., Yang, W., Liu, K., Pawlik, K. M., Erdjument-Bromage, H., Tempst, P., Lee, Y., Min, J., Townes, T. M. & Wang, H. PRC2 complexes with JARID2, MTF2, and esPRC2p48 in ES cells to modulate ES cell pluripotency and somatic cell reprogramming. Stem Cells 29 , 229–240 (2011). Additional Declarations There is NO Competing Interest. Supplementary Files FigS12flat.pdf Figure S1. Identification of EPOP as syndapin I interaction partner (A,B) Immunoblotting analyses of coprecipitation experiments with purified fusion proteins of syndapin I (GST-SdpI SH3 ; immobilized) and EPOP (TrxHis-EPOP 222-369 (A) and TrxHis-EPOP 52-369 (B) clearly proving a direct interaction. Green arrows mark position of TrxHis-EPOP 222-369 (A) and TrxHis-EPOP 52-369 (B), respectively. Red arrows mark position of TrxHis not precipitated by GST-SdpI SH3 . GST and TrxHis served as specificity controls. Anti-EPOP immunoblots of input and eluates are repeated from Fig. 1C for comparison. (C) Immunoblotting analyses of coprecipitation attempts of full-length GFP-EPOP from cellular extracts with different immobilized GST-fusion proteins of syndapin I and GST (control), respectively, demonstrating that the syndapin I SH3 domain is required and sufficient for the interaction with full-length EPOP. Anti-GFP immunoblots of input and eluates are repeated from Fig. 1D for comparison. FigS24flat.pdf Figure S2. EPOP localizes to nucleoli in neurons (A) In Western blot analyses of lysates from transfected HEK293 with anti-EPOP (upper panel) and anti-GFP antibodies (lower panel). The affinity-purified rabbit-anti-EPOP antibody specifically detects expressed tagged and untagged mouse and rat EPOP proteins, but not GFP coexpressed as a control. The anti-GFP immunoblot highlights GFP and GFP-tagged EPOP proteins for comparison. White lines represent omitted lanes. (B) Immunoblot analyses with decreasing amounts of TrxHis-EPOP 52-369 and affinity-purified rabbit-anti-EPOP antibodies. The anti-EPOP antibodies specifically detect 0.001 µg of recombinant, purified mouse TrxHis-EPOP 52-369 and recognizes endogenous EPOP of the expected molecular size (47 kD; compare untagged overexpressed EPOP in A) in 25 µg of rat brain extracts. (C) Immunostaining of primary hippocampal rat cultures (DIV18) showed that endogenous EPOP (green in merges) is present in nucleoli highlighted by anti-fibrillarin staining (red in merges), nucleoplasm overlapping with DAPI (not included in merges) and somatodendritic cytoplasm, which was stained with antibodies against MAP2 (blue in merges). Bar, 10 µm. FigS31flat.pdf Figure S3. EPOP immunoreactivity during neuronal development Endogenous staining of primary hippocampal rat neurons showed an increasing somatodendritic localization of EPOP (green in merges) during neuronal development, overlapping with anti-syndapin I immunoreactivity (red in merges). The nuclei were stained with DAPI (omitted from the merges) and the dendrites with an anti-MAP2 antibody (blue in merges). Bars, 20 µm. Cite Share Download PDF Status: Under Review 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-9117199","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":612520160,"identity":"9d1f3501-1266-461d-a797-d0c86b0236c2","order_by":0,"name":"Michael Kessels","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0001-5967-0744","institution":"Jena University Hospital - Friedrich Schiller University Jena","correspondingAuthor":true,"prefix":"","firstName":"Michael","middleName":"","lastName":"Kessels","suffix":""},{"id":612520161,"identity":"073f1a18-db2a-43d7-8af2-fe68f7e64fb9","order_by":1,"name":"Jacqueline Dömming","email":"","orcid":"","institution":"Jena University Hospital - Friedrich Schiller University Jena","correspondingAuthor":false,"prefix":"","firstName":"Jacqueline","middleName":"","lastName":"Dömming","suffix":""},{"id":612520162,"identity":"a667e4a0-4c52-4bf2-85b9-24eafc125c31","order_by":2,"name":"Regina Dahlhaus","email":"","orcid":"","institution":"Jena University Hospital - Friedrich Schiller University Jena","correspondingAuthor":false,"prefix":"","firstName":"Regina","middleName":"","lastName":"Dahlhaus","suffix":""},{"id":612520163,"identity":"0f4f8f6c-9c50-47d8-a1a6-4fa2e599ca53","order_by":3,"name":"Nicole Koch","email":"","orcid":"","institution":"Jena University Hospital - Friedrich Schiller University Jena","correspondingAuthor":false,"prefix":"","firstName":"Nicole","middleName":"","lastName":"Koch","suffix":""},{"id":612520164,"identity":"3fbaa235-d849-4c05-959a-8b8e0816df8a","order_by":4,"name":"Britta Qualmann","email":"","orcid":"https://orcid.org/0000-0002-5743-5764","institution":"Jena University Hospital - Friedrich Schiller University Jena","correspondingAuthor":false,"prefix":"","firstName":"Britta","middleName":"","lastName":"Qualmann","suffix":""}],"badges":[],"createdAt":"2026-03-13 17:50:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9117199/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9117199/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105704531,"identity":"109f4a39-6bfc-4f32-8445-a7bb8c8ba3a0","added_by":"auto","created_at":"2026-03-30 06:44:36","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":517305,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of EPOP as syndapin I interaction partner\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Schematic representation of six independent Y2H clones encoding for C-terminal fragments of the mouse protein EPOP (NP_780541.2) isolated from an embryonal mouse brain library by Y2H-screening with full-length syndapin I as bait. (B) Verification of the syndapin I interaction using reisolated plasmids of the prey clones #320 (aa52-369) and #327 (aa222-369) by reporter gene activity assessment (growth on quadruple drop-out plates and β-Gal activity, respectively). Note that the experiments demonstrate an interaction of syndapin I, but not of syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e, with C-terminal fragments of EPOP. pGBTK7 served as negative control for the bait vectors. (C) Immunoblotting analyses of coprecipitation experiments with purified fusion proteins of syndapin I (GST-SdpI\u003csup\u003eSH3\u003c/sup\u003e; immobilized) and EPOP (TrxHis-EPOP\u003csup\u003e52-369\u003c/sup\u003e) proving a direct interaction. Green arrows mark position of TrxHis-EPOP\u003csup\u003e52-369\u003c/sup\u003e, red arrows mark position of TrxHis not precipitated by GST-SdpI\u003csup\u003eSH3\u003c/sup\u003e. GST and TrxHis served as specificity controls. For additional display of supernatants and of the anti-GST immunoblots of the eluates, see Fig. S1. (D) Immunoblotting analyses of coprecipitation attempts of full-length GFP-EPOP from cellular extracts with different immobilized GST-fusion proteins of syndapin I and GST (control), respectively, demonstrating that the syndapin I SH3 domain is required and sufficient for the interaction with full-length EPOP. For additional immunoblot of the supernatants, see Fig. S1. (E) Immunoblotting analyses of coimmunoprecipitation experiments from lysates of HEK293 cells coexpressing Flag-EPOP together with GFP-syndapin I (GFP-SdpI) or GFP as negative control. Specific coimmunoprecipitation of EPOP/syndapin I complexes were obtained by anti-Flag antibodies but not by control IgGs. (F-I) Recruitment experiments in COS-7 cells showing that mito-mCherry-EPOP (G,I; red in merges) is able to interact with Xpress-syndapin I (F,G; green in merges). White arrows in G highlight coaccumulations of syndapin I and EPOP at mitochondria (G; yellow in merge). Xpress-syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e (H,I; green in merges) was not recruited to mito-mCherry-EPOP (I). Mito-mCherry alone (F,H; red in merges) did not recruit any of the syndapin I proteins to mitochondria. Mitochondria were labeled with MitoTracker\u003csup\u003e®\u003c/sup\u003e DeepRed 633 (magenta in merges). DNA was stained using DAPI (blue in merges). Bars, 20 µm.\u003c/p\u003e","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/2970853baedaf6eeb7d150e3.jpg"},{"id":105704537,"identity":"42e83a73-cccc-4c32-a395-7e584571a259","added_by":"auto","created_at":"2026-03-30 06:44:36","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1048331,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEPOP is a neuronal protein with both nuclear and cytoplasmic localization\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Immunoblot analysis of fractions of adult mouse cortex homogenates (40 µg each) showing anti-EPOP bands in the range of 45-50 kD in both nuclear pellet (red arrow) and supernatant fraction (green arrow). Anti-laminB1 and anti-GAPDH immunoreactivities identify nuclear and cytoplasmic fractions, respectively. In contrast to EPOP, syndapin I was not detected in the pellet fraction. (B) Immunostaining of a sagittal adult mouse brain section with antibodies against EPOP and MAP2 showing the rather ubiquitous expression of EPOP in the mouse brain. Bar, 1 mm. (C) Endogenous EPOP was detected in various brain regions of adult rats by immunoblot analysis of homogenates (50 µg each). (D) Immunostaining of the hippocampal CA1 region in adult mouse brain slices revealed predominant expression of EPOP (green in merge) in neurons, highlighted by anti-MAP2 staining (red in merge) compared to astrocytes identified by anti-GFAP labeling (blue in merge). The nuclei of all cells were stained with DAPI (excluded in the merge). Bottom panels show enlargements of the boxed area. Bar, 50 µm. (E,F) In both adult mouse brain slices (E) and DIV18 primary hippocampal rat cultures (F), endogenous EPOP (green in merges) was detected in the somatodendritic but not in the axonal compartment. The dendrites and soma were labeled by anti-MAP2 staining (red in merges) and the axons by anti-SMI312 staining (blue in merges). The nuclei were stained using DAPI (omitted in the merges). Bars, 50 µm (E) and 20 µm (F).\u003c/p\u003e","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/0a31eeec24c8c2476a50f941.jpg"},{"id":105704532,"identity":"b2151533-7022-434e-84fe-81b54690dc55","added_by":"auto","created_at":"2026-03-30 06:44:36","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":476546,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSyndapin I deficiency diminishes the extranuclear localization of EPOP in neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A,B) Coexpression of syndapin I and GFP-EPOP in COS-7 cells (A) resulted in an increased cytoplasmic localization of EPOP compared to single expression of GFP-EPOP (B). n=71 (GFP-EPOP) and n=67 (GFP-EPOP + syndapin I) cells from 2 independent experiments were evaluated. The mean values of the intensity quotient (mean intensity in the nucleus/mean intensity in the cytoplasm) are given as a deviation from GFP-EPOP with the respective standard error. Mann-Whitney test, p=0.0003 (***). (C) Immunoblotting with guinea pig anti-syndapin I antibodies showed that endogenous syndapin I was coprecipitated with GST-EPOP\u003csup\u003e222-369 \u003c/sup\u003ebut not GST-EPOP\u003csup\u003e1-169 \u003c/sup\u003eor GST from mouse brain extracts.\u003cstrong\u003e \u003c/strong\u003e(D) Representative images of hippocampal \u003cem\u003esyndapin I\u003c/em\u003e WT and KO neurons immunostained at DIV17. (E) Modeled VOI-Rs of the individual channels and look-up table (LUT) display (Fire) by EPOP. Enlargements represent boxed areas of the overview image. (F-H) Quantitative analyses of EPOP intensity ratios in the different compartments. Related to the anti-EPOP immunoreactivity of the whole cell, a significant increase of EPOP immunoreactivity was seen in the whole nucleus (F) and in the nucleoplasm (G) of \u003cem\u003esyndapin I\u003c/em\u003e KO neurons compared to WT neurons. Also in the nuclear-extranuclear ratio, a clear shift from EPOP towards nuclear compartment was seen in \u003cem\u003esyndapin I\u003c/em\u003e KO neurons when compared to WT neurons (H). A total of 48 (WT) and 46 (\u003cem\u003esyndapin I\u003c/em\u003e KO) neurons from 3 independent preparations were evaluated. The mean values of the difference from WT neurons with the respective standard error are given. Unpaired t-test: **; p\u0026lt;0.01; *, p\u0026lt;0.05. Bars, 10 µm.\u003c/p\u003e","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/b036bdd9609b46b8488794da.jpg"},{"id":105704541,"identity":"a05b665a-7d02-43fc-9147-e03dbe25c071","added_by":"auto","created_at":"2026-03-30 06:44:37","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":626491,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEPOP is localized to dendritic spines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A,B) Immunoblot analysis of mouse brain homogenates (A) and lysed cortical rat primary cultures (B), respectively, at different developmental stages indicated an increase in EPOP expression (red and green arrows) during neuronal development comparable to that of the synaptic proteins syndapin I, synapsin1 and ProSAP1. An additional band at 40 kD (black arrow) showed an inverse profile. 25 µg of total protein each were loaded (A) and volumes were adjusted to ß-tubulin and laminB1 signals (B), respectively. (C) Immunostaining of primary hippocampal rat neurons (DIV18) revealed that endogenous EPOP (green in merges) colocalized with homer1 (red in merges) in the postsynapse (arrowheads). A comparison with anti-MAP2 (blue in merges) and DAPI (omitted from the merge) clarified the somatodendritic and nuclear localization of EPOP, respectively. Bars, 10 µm. Magnifications show enlargements of the boxed areas. (D) In mouse primary hippocampal neurons (DIV17), EPOP (green in merges) colocalized in postsynaptic spines (arrowheads) with PSD95 (blue in merges). Additional colocalization with syndapin I (red in merges) was observed. Bars (C,D), 10 µm and 2 µm (magnifications).\u003c/p\u003e","description":"","filename":"fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/500a35818eeef3eab5d84f41.jpg"},{"id":105704535,"identity":"01d49e7b-c487-4bad-b469-9aca9d4f94c8","added_by":"auto","created_at":"2026-03-30 06:44:36","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":420992,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEPOP deficiency impairs mushroom-type dendritic spines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Quantitative immunoblot analysis of lysates from HEK293 cells transfected with scr-RNAi_GFP_IRES_EPOP, EPOP-RNAi-2_GFP_IRES_EPOP and EPOP-RNAi-5_GFP_IRES_EPOP, respectively. After 48 h, the depletion of rat EPOP with EPOP-RNAi-5 was highest. 6 independent experiments; Shapiro Wilk normality assessment and one-way ANOVA/Bonferroni’s test: *, p\u0026lt;0.5; **, p\u0026lt;0.01; ***, p\u0026lt;0.001. (B,C) Representative immunoblot and quantitative evaluation of cortical rat primary cultures (DIV17) treated with anisomycin (7.5 µM final) showing a significant reduction in endogenous EPOP protein levels after 12 h and 24 h compared to DMSO-treated cells. The measurement of the expression of EPOP under anisomycin normalized to ß-tubulin was time-related to the respective DMSO controls. 4 independent experiments, two-way ANOVA: interaction, p = 0.0005; column factor, p = 0.0001; series factor, p = 0.0005; ***, p\u0026lt;0.001. (D) Representative MIP images of the mCherryF signal from primary hippocampal rat neurons transfected (DIV16+2) as indicated and the respective filament modeled using Imaris. Bars, 2 µm. (E-G) Quantitative evaluations of dendritic spine parameters. The density of all dendritic spines (E) and of mushroom-shaped dendritic spines (F) was significantly reduced in EPOP and syndapin I RNAi conditions compared to scrambled RNAi (scr-RNAi) (except the total density in EPOP-RNAi-5). The head width of the mushroom-shaped dendritic spines was significantly reduced in both EPOP RNAi conditions and upon syndapin I RNAi (G). A total of 29 (syndapin I-RNAi) or 30 (src-RNAi, EPOP-RNAi-2, EPOP-RNAi-5) dendrite sections per condition (E,F) as well as 1369 (scr-RNAi), 1052 (EPOP-RNAi-2), 1005 (EPOP-RNAi-5) and 756 (syndapin I-RNAi) mushroom spines (G) from 3 independent experiments were evaluated. The mean values of the deviation from the control with the respective standard error are given, normalized for the individual experiments. One-way ANOVA/Bonferroni’s (p = 0.0134 in E) and the Kruskal Wallis/Dunn’s (p \u0026lt; 0.0001 in F, p \u0026lt; 0.0001 in G), respectively; *, p\u0026lt;0.05; **, p\u0026lt;0.01; ***, p\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/9631d632b80700d685b7a0ed.jpg"},{"id":105704538,"identity":"f622e1b3-3a3f-4137-ad99-b176a8656bb2","added_by":"auto","created_at":"2026-03-30 06:44:37","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":393329,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEPOP loss-of-function phenotype on dendritic mushroom spines is rescued by reexpression of RNAi-resistant EPOP*\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A,B) Representative immunoblot (A) and quantitative evaluations (B) of the effects of coexpressed EPOP-RNAi-5 in comparison to scr-RNAi of WT rat EPOP and RNAi-resistant mouse EPOP*. EPOP RNAi-5 was able to deplete WT rat EPOP but not RNAi-resistant mouse EPOP* coexpressed in HEK293 cells for 48 h. n=6 independent experiments. Kolmogorov–Smirnov normality test; one-way ANOVA/Bonferroni’s: ***, p\u0026lt;0.001. (C) Representative MIP images of the mCherryF-signal of primary hippocampal rat neurons (DIV16+2) transfected as indicated and their corresponding filaments modeled with Imaris. Bars, 2 µm. (D-F) Quantitative evaluations of dendritic spine parameters. While significant reductions were observed for EPOP-RNAi-5_IRES in the density (E) as well as in the head width of mushroom-shaped dendritic spines (F), no significant difference to the control was observed when co-expressing the RNAi-resistant EPOP* (rescue condition proving the specificity of the identified EPOP loss-of-function phenotypes). n=31 (EPOP-RNAi-5_IRES) or 32 (scr-RNAi_IRES, EPOP-RNAi-5_IRES_EPOP*) dendrite sections per condition (D,E) and 2076 (scr-RNAi_IRES), 1495 (EPOP-RNAi-5_IRES) or 2224 (EPOP-RNAi-5_IRES_EPOP*) mushroom spines (F) from 2 independent experiments. The mean values of the deviation from the control with standard error are given, normalized to the respective experiment. One-way ANOVA/Bonferroni’s (p = 0.0154 in D, p \u0026lt; 0.0001 in E) and Kruskal-Wallis/Dunn’s (p \u0026lt; 0.0018 in F), respectively; *, p\u0026lt;0.05; **, p\u0026lt;0.01; ***, p\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/6b65d0a7355df23bc4aa97cd.jpg"},{"id":105704540,"identity":"e54dcaa2-4653-4b1e-941b-cec59540a6d9","added_by":"auto","created_at":"2026-03-30 06:44:37","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":178282,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDepletion of EPOP significantly affects homer1 in dendritic spines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative images of homer1 VOI-Rs of a dendritic segment of primary hippocampal rat neurons (DIV16+2) transfected with scr-RNAi, EPOP-RNAi-5 and syndapin I-RNAi, respectively, after reconstruction in Imaris. Bars, 2 µm. (B-E) Quantitative evaluations of the VOI-Rs of homer1. Both EPOP and syndapin I depletion resulted in a significant relative reduction of homer1 VOI-Rs in all dendritic spines (B) and particularly in mushroom-shaped spines (C). In the remaining homer1 VOI-Rs, the mean intensity was significantly lower upon EPOP and syndapin I RNAi compared to control (D). EPOP and syndapin I depletion resulted in reduced volume of the homer1 VOI-Rs (E). n=30 (scr-RNAi B-F; EPOP-RNAi-5 B,C), 29 (syndapin I-RNAi B,C) and 28 (EPOP-RNAi-5 D,E syndapin I-RNAi D,E) dendrite sections per condition from 3 independent experiments. The mean values of the deviation from the control with the standard error are given, normalized for the respective experiment. One way ANOVA/Bonferroni’s (p \u0026lt; 0.0001 in C, p = 0.0002 in D, p \u0026lt; 0.0001 in E) and Kruskal-Wallis/Dunn’s (p \u0026lt; 0.0001 in B left), respectively; **, p\u0026lt;0.01; ***, p\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"fig7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/d5b6d6fc5ee2bd419e53b71f.jpg"},{"id":105752685,"identity":"e8710f0e-3c61-4232-bd6c-d01574789c3b","added_by":"auto","created_at":"2026-03-30 16:04:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4992189,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/0536c5c1-061a-4ee8-9e05-e2547bad505c.pdf"},{"id":105729222,"identity":"15b97e03-77ae-4cf2-b28b-6b5556c35af9","added_by":"auto","created_at":"2026-03-30 11:13:49","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":961805,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure S1. Identification of EPOP as syndapin I interaction partner\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A,B) Immunoblotting analyses of coprecipitation experiments with purified fusion proteins of syndapin I (GST-SdpI\u003csup\u003eSH3\u003c/sup\u003e; immobilized) and EPOP (TrxHis-EPOP\u003csup\u003e222-369 \u003c/sup\u003e(A)\u003csup\u003e \u003c/sup\u003eand TrxHis-EPOP\u003csup\u003e52-369 \u003c/sup\u003e(B) clearly proving a direct interaction. Green arrows mark position of TrxHis-EPOP\u003csup\u003e222-369 \u003c/sup\u003e(A)\u003csup\u003e \u003c/sup\u003eand TrxHis-EPOP\u003csup\u003e52-369\u003c/sup\u003e (B), respectively. Red arrows mark position of TrxHis not precipitated by GST-SdpI\u003csup\u003eSH3\u003c/sup\u003e. GST and TrxHis served as specificity controls. Anti-EPOP immunoblots of input and eluates are repeated from Fig. 1C for comparison. (C) Immunoblotting analyses of coprecipitation attempts of full-length GFP-EPOP from cellular extracts with different immobilized GST-fusion proteins of syndapin I and GST (control), respectively, demonstrating that the syndapin I SH3 domain is required and sufficient for the interaction with full-length EPOP. Anti-GFP immunoblots of input and eluates are repeated from Fig. 1D for comparison.\u003c/p\u003e","description":"","filename":"FigS12flat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/8a5f55b813e4be29857182ac.pdf"},{"id":105704534,"identity":"5916c62f-d046-40b7-8c20-700376ca33a0","added_by":"auto","created_at":"2026-03-30 06:44:36","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":945965,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure S2. EPOP localizes to nucleoli in neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) In Western blot analyses of lysates from transfected HEK293 with anti-EPOP (upper panel) and anti-GFP antibodies (lower panel). The affinity-purified rabbit-anti-EPOP antibody specifically detects expressed tagged and untagged mouse and rat EPOP proteins, but not GFP coexpressed as a control. The anti-GFP immunoblot highlights GFP and GFP-tagged EPOP proteins for comparison. White lines represent omitted lanes. (B) Immunoblot analyses with decreasing amounts of TrxHis-EPOP\u003csup\u003e52-369 \u003c/sup\u003eand affinity-purified rabbit-anti-EPOP antibodies. The anti-EPOP antibodies specifically detect 0.001 µg of recombinant, purified mouse TrxHis-EPOP\u003csup\u003e52-369 \u003c/sup\u003eand recognizes endogenous EPOP of the expected molecular size (47 kD; compare untagged overexpressed EPOP in A) in 25 µg of rat brain extracts. (C) Immunostaining of primary hippocampal rat cultures (DIV18) showed that endogenous EPOP (green in merges) is present in nucleoli highlighted by anti-fibrillarin staining (red in merges), nucleoplasm overlapping with DAPI (not included in merges) and somatodendritic cytoplasm, which was stained with antibodies against MAP2 (blue in merges). Bar, 10 µm.\u003c/p\u003e","description":"","filename":"FigS24flat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/304230b6c5499e627fc7de7e.pdf"},{"id":105729246,"identity":"aeb1e302-54b4-448a-9be0-f6591d98c9f5","added_by":"auto","created_at":"2026-03-30 11:14:01","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":1593661,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure S3. EPOP immunoreactivity during neuronal development\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEndogenous staining of primary hippocampal rat neurons showed an increasing somatodendritic localization of EPOP (green in merges) during neuronal development, overlapping with anti-syndapin I immunoreactivity (red in merges). The nuclei were stained with DAPI (omitted from the merges) and the dendrites with an anti-MAP2 antibody (blue in merges). Bars, 20 µm.\u003c/p\u003e","description":"","filename":"FigS31flat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9117199/v1/0dc688f3988ca47b4283e179.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Identification of moonlighting, extranuclear and syndapin I-related functions of EPOP in mature neurons","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe polycomb repressive complex 2 (PRC2) represents an important chromatin regulatory complex and functions by repressing the transcription of diverse developmental genes. PRC2 consists of a core complex responsible for catalyzing both di- and trimethylation of histone H3 at lysine 27 and various associated components that modulate its function in different ways (Fischer et al., 2022; Aguilar et al., 2025). In the nervous system, PRC2 has been reported to play important roles in neuronal identity, proliferation and differentiation of neural stem and progenitor cells as well as in gliogenesis. Mutations or dysregulations of PRC2 components are frequently associated with neurological diseases (Liu et al., 2018).\u003c/p\u003e \u003cp\u003eEPOP (Elongin BC and Polycomb repressive complex 2-associated protein; also named C17orf96, esPRC2p48, E130012A19Rik) is a mammalian-specific PRC2-associated factor (Zhang et al., 2011; De Cegli et al., 2013; Smits et al., 2013; Alekseyenko et al., 2014). Whereas EPOP was shown to enhance the methyltransferase activity of PRC2 in vitro (Behringer et al., 2016; Zhang et al., 2011), EPOP deletion resulted in an elevated chromatin association of PRC2 and increased repression of PRC2 target genes suggesting that EPOP represses PRC2 function in vivo (Liefke \u0026amp; Shi, 2015; Behringer et al., 2016; Liefke et al., 2016; Healy et al., 2019; Granat et al., 2026). Recently, EPOP was shown to restrict the targeting of specifically the PCR2.1 complex to chromatin by disrupting the dimeric architecture of the enzyme complex (Gong et al., 2026). EPOP furthermore also interacts with the heterodimer Elongin BC (Behringer et al., 2016; Liefke et al., 2016), which plays a role in transcription elongation and protein turnover (Okumara et al., 2012). The cooperative of EPOP with Elongin BC to increase gene transcription at actively transcribed genes may additionally involve an EPOP interaction with the deubiquitinase USP7 (Liefke et al., 2016). Analyses of \u003cem\u003eepop\u003c/em\u003e knockout (KO) mice, however, revealed no major defects when several of the PRC2-related developmental processes were examined, but \u003cem\u003eepop\u003c/em\u003e KO mice exhibited posterior homeotic transformations of the axial skeleton and a shift of the anterior boundary of the expression of certain \u003cem\u003ehox\u003c/em\u003e genes (Mocavini et al., 2025). All these observations revealed function for EPOP as a nuclear scaffold protein in early cellular differentiation.\u003c/p\u003e \u003cp\u003eStrikingly, we here identify EPOP as a binding partner for the extranuclear, membrane-binding F-BAR protein syndapin I (also called PACSIN1) (Qualmann et al., 1999; Dharmalingam et al. 2009) and reveal a crucial role for the two interacting proteins in dendritic spines of mature neurons.\u003c/p\u003e \u003cp\u003eSyndapin I is a neuron-enriched protein that has important roles in both neuromorphogenesis and synaptic functions. In presynapses, syndapin I is crucial for activity-dependent membrane trafficking and shaping of synaptic vesicles (Anggono et al. 2006; Koch et al. 2011). In the postsynaptic compartment, syndapin I was demonstrated to be involved in dendritic spine formation and organization (Schneider et al. 2014) and to regulate the availabilities of both AMPA- and NMDA-type glutamate receptors during basal synaptic activity and during synaptic plasticity, as \u003cem\u003esyndapin I\u003c/em\u003e KO altered postsynaptic actin dynamics, synaptic glutamate receptor clustering, as well as the mobility and internalization of AMPA receptors (Koch et al., 2020). As a consequence, \u003cem\u003esyndapin I\u003c/em\u003e KO mice displayed schizophrenia-like behavior and epileptic seizures (Koch et al., 2011; Koch et al., 2020).\u003c/p\u003e \u003cp\u003eBiochemical, immunocytochemical and immunohistochemical analyses revealed that EPOP, in addition to its nuclear localization, is a somatodendritic protein in mature neurons. Studies in \u003cem\u003esyndapin I\u003c/em\u003e KO neurons uncovered that this extranuclear presence of EPOP depended on its direct binding partner, syndapin I. Loss-of-function of EPOP in mature hippocampal neurons negatively impacted the density, morphology and postsynaptic-density composition of mushroom-type dendritic spines. Beyond EPOP\u0026rsquo;s PRC2-related role in cell differentiation and early development, the observed phenocopy of the postsynaptic phenotypes of syndapin I knockdown in neurons strongly suggest thus far unrecognized, extranuclear functions of EPOP in mature, postmitotic cells, which already are fully integrated into cellular networks.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEPOP interacts directly with the F-BAR protein syndapin I via an SH3 domain-dependent mechanism\u003c/h2\u003e \u003cp\u003ePrevious studies have identified the F-BAR protein syndapin I as a crucial postsynaptic coordinator in the formation of excitatory synapses and dendritic spines. Syndapin I-enriched membrane nanodomains thereby appear to represent important spatial cues and organizing platforms, shaping dendritic membrane areas into synaptic compartments (Schneider et al., 2014). To deepen the molecular understanding of this crucial role of syndapin I, we performed yeast two-hybrid (Y2H) screens with syndapin I as bait.\u003c/p\u003e \u003cp\u003eInterestingly, several clones identified encoded for six independent C-terminal fragments of the mouse protein\u0026ensp;elongin BC and polycomb repressive complex 2 associated protein (EPOP; NP_780541.2), with the longest Y2H clone #320 representing aa52-369 and the shortest C-terminal fragment (Y2H clone #327) spanning aa222-369 of mouse EPOP (Fig.\u0026nbsp;1A). Retransformation of the isolated EPOP prey plasmids with syndapin I bait plasmids confirmed the specificity of the Y2H hits by both growth on drop-out plates and β-galactosidase activity. Reporter gene assessments using syndapin I\u003csup\u003e∆SH3\u003c/sup\u003e as bait furthermore revealed that the syndapin I SH3 domain was required for the interaction (Fig.\u0026nbsp;1B).\u003c/p\u003e \u003cp\u003eReconstitutions with purified proteins demonstrated that a GST-fusion protein of the syndapin I SH3 domain specifically coprecipitated different TrxHis-fusion proteins of EPOP (Fig.\u0026nbsp;1C and Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e A,B). The syndapin I/EPOP interaction therefore is direct. Coprecipitation analyses with immobilized recombinant GST-syndapin I and deletion mutants thereof together with GFP-EPOP (aa1-369) demonstrated the interaction of the two full-length proteins and furthermore showed that the SH3 domain of syndapin I is both required and sufficient for the association with EPOP (Fig.\u0026nbsp;1D and Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCoi\u003c/strong\u003e \u003cp\u003emmunoprecipitation experiments demonstrated the in vivo-relevance of the EPOP interaction with syndapin I. GFP-syndapin I was coimmunoprecipitated with Flag-tagged EPOP in a specific manner (Fig.\u0026nbsp;1E). Also in intact cells, EPOP and syndapin I formed protein complexes as demonstrated by specific reconstitution experiments at outer mitochondrial membranes (Fig.\u0026nbsp;1F-I). In contrast to the recruitment of wild-type syndapin I, mito-mCherry-EPOP-decorated mitochondria did not recruit syndapin I\u003csup\u003eSH3\u003c/sup\u003e (Fig.\u0026nbsp;1I). Thus, also in living, intact cells, a specific association between EPOP and syndapin I can be observed and is critically mediated by the SH3 domain of syndapin I (Fig.\u0026nbsp;1F-I).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eThe subcellular distribution of EPOP is not restricted to the nucleus but anti-EPOP immunoreactivity can also be observed in the neuronal dendritic arbor\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo our knowledge, functions described for EPOP have so far been restricted to the nucleus mainly mediated through its association with the polycomb repressive complex 2, which is responsible for histone methylation, and the transcription elongation factor Elongin BC (Bartke et al., 2010; Zhang et al., 2011; Liefke et al., 2016; Beringer et al., 2016). Our identification of syndapin I, a specifically neuron-enriched member of the syndapin subfamily of F-BAR domains proteins, which had only been detected at extranuclear locations within cells, as binding partner of EPOP raised the important question whether the localization of EPOP would be restricted to nuclear compartments or whether EPOP might instead be able to act as a nucleocytoplasmic effector.\u003c/p\u003e \u003cp\u003eBiochemical fractionation of mouse brain homogenates analyzed with antibodies raised and affinity-purified against mouse EPOP (aa52-369) (for anti-EPOP antibody characterization see Supplementary Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e A,B) showed that endogenous EPOP was not only detected in pellet fractions containing cell nuclei, as demonstrated by efficient labeling with antibodies against the nuclear marker laminB1, but also in supernatant fractions devoid of anti-laminB1 signals but instead showing enrichment of the cytoplasmic protein GAPDH (Fig.\u0026nbsp;2A). Anti-EPOP immunoreactivity was detected as two bands with slightly different apparent molecular weights. EPOP was observed in form of a 50 kD band that was nucleus-enriched and a more abundant 48 kD band, which was mainly detected in the postnuclear supernatant (Fig.\u0026nbsp;2A). Especially the 50 kD band runs significantly higher than the calculated molecular weight of EPOP (39 kD). Yet, both bands were also visible when overexpressed untagged EPOP was detected by anti-EPOP antibodies. Both bands thus were EPOP-specific, i.e. represent different forms of EPOP originating from the same mRNA (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eA). In line, EPOP is not predicted to have splice variants but is encoded by a single exon. Syndapin I was clearly detectable in the supernatant but not in the pellet fraction (Fig.\u0026nbsp;2A). These experiments thus argued that EPOP/syndapin I complexes are cytosolic and therefore represent a thus far unnoticed function of EPOP that is different from EPOPs known association with the polycomb repressive complex 2 in the nucleus.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSyndapin I exhibits high expression levels in various regions of the brain and had been shown to play important roles in hippocampal and cortical neurons (Dharmalingam et al., 2009; Koch et al., 2011; Schneider et al., 2014; Koch et al., 2020). We therefore next analyzed the expression pattern of EPOP in adult brains by both Western blot and immunofluorescence analyses. Immunocytochemistry of sagittal sections of adult mouse brains detected high levels of endogenous EPOP in various regions including the striatum, hippocampus, cortex and cerebellum (Fig.\u0026nbsp;2B).\u003c/p\u003e \u003cp\u003eThis rather wide-spread expression of EPOP in the brain was corroborated by Western blot analyses of tissue material from various brain regions. EPOP protein expression in the brain overlapped significantly with the distribution of syndapin I (Fig.\u0026nbsp;2C).\u003c/p\u003e \u003cp\u003eTo unequivocally address the localization of endogenous EPOP in neuronal cells, EPOP immunoreactivity was analyzed in comparison to appropriate markers in mouse brain slices. Endogenous EPOP colocalized with MAP2, an exclusively neuronal expressed protein, and not with GFAP, a typical glial marker protein for astrocytes, as demonstrated for the CA1 region of the hippocampus (Fig.\u0026nbsp;2D). Thus, endogenous EPOP is preferably expressed in neurons.\u003c/p\u003e \u003cp\u003eIn neurons, EPOP did not only show a nuclear localization and a particularly high presence in nucleoli (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eC), but also a cytoplasmic localization. This property of EPOP was observed both in CA1 and the dentate gyrus (Fig.\u0026nbsp;2E).\u003c/p\u003e \u003cp\u003eStrikingly, no colocalization of endogenous EPOP with SMI312, a pan-axonal neurofilament marker, was detectable in both mouse brain sections (Fig.\u0026nbsp;2E) and in primary hippocampal rat cultures (Fig.\u0026nbsp;2F). In contrast, EPOP showed clear overlay with the dendritic marker MAP2 (Fig.\u0026nbsp;2D-F). Together, these observations suggested that EPOP\u0026rsquo;s extranuclear functions may predominantly be of cytoplasmic and dendritic rather than axonal nature.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eThe EPOP binding partner syndapin I is important for the nucleocytoplasmic distribution of EPOP\u003c/h3\u003e\n\u003cp\u003eOur data raised the important question what determines the localization of EPOP to various subcellular compartments of neurons and how and to what extent these distinct EPOP localizations are dependent on the respective binding partners of EPOP. Importantly, the newly identified EPOP binding partner syndapin I influenced EPOP\u0026rsquo;s nucleocytoplasmic distribution. Ectopic overexpression of syndapin I in COS-7 cells led to elevated levels of GFP-EPOP in the cytoplasm (Fig.\u0026nbsp;3A). Quantitative analyses showed that the ratio of nuclear to cytoplasmic EPOP dropped by about\u0026thinsp;\u0026minus;\u0026thinsp;32% when syndapin I was present (Fig.\u0026nbsp;3A,B). The induced localization change of EPOP from the nuclear to the cytoplasmic compartment by syndapin I presence suggested that syndapin I might act as an extranuclear anchor for EPOP.\u003c/p\u003e \u003cp\u003eCoprecipitation experiments indeed demonstrated that endogenous syndapin I was specifically coprecipitated from mouse brain extracts by immobilized EPOP\u003csup\u003e222\u0026ndash;369\u003c/sup\u003e (Fig.\u0026nbsp;3C). Interestingly, this was not the case when an N terminal fragment of EPOP was used suggesting that the endogenous syndapin I protein from brain interacted with the rather proline-rich C terminal half of EPOP (Fig.\u0026nbsp;3C).\u003c/p\u003e \u003cp\u003eWe next analyzed whether syndapin I was indeed important for controlling the subcellular localization of endogenous EPOP in neurons. For this purpose, intensity ratios of the immunosignals of EPOP measured in different subcellular compartments in primary hippocampal neurons prepared from wild-type (WT) and \u003cem\u003esyndapin I\u003c/em\u003e KO neurons (Koch et al., 2011) were compared (Fig.\u0026nbsp;3D,E). When the ratios of nuclear EPOP immunoreactivity was related to the whole cell EPOP immunoreactivity, significant nuclear accumulation of EPOP was detected in \u003cem\u003esyndapin I\u003c/em\u003e KO neurons compared to WT neurons (Fig.\u0026nbsp;3F). A significant increase was also observed in the nucleoplasm of syndapin I-deficient neurons (Fig.\u0026nbsp;3G).\u003c/p\u003e \u003cp\u003eIn order to further corroborate these findings and to explicitly investigate the cytoplasmic pool of EPOP, we also determined the nuclear anti-EPOP immunolabeling intensities related to the respective extranuclear, i.e. cytoplasmic, pool of EPOP immunoreactivity. Also this ratio showed a significant increase (+\u0026thinsp;11%; Fig.\u0026nbsp;3H).\u003c/p\u003e \u003cp\u003eThus, syndapin I influences the localization of EPOP at the endogenous level. In the presence of syndapin I, EPOP was located relatively less frequently in the nucleus.\u003c/p\u003e\n\u003ch3\u003eEPOP is localized to dendritic spines\u003c/h3\u003e\n\u003cp\u003eOur studies observed a high expression of EPOP in the adult neuronal tissue and in mature primary cultures. So far, EPOP had been mainly described as an important protein in embryonic stem cells (Zhang et al. 2011, Liefke \u0026amp; Shi 2015). Investigating EPOP protein expression levels in brain homogenates of mice of different age stages, both the 50 kD band corresponding to nuclear-enriched EPOP (red arrow in Fig.\u0026nbsp;4A; compare Fig.\u0026nbsp;2A) and corresponding to the 47 kD band of EPOP reflecting the cytoplasmic EPOP (green arrow in Fig.\u0026nbsp;4A; compare Fig.\u0026nbsp;2A) could be detected in the E16 embryonic stage. During development, the 50 kD band reflecting nuclear EPOP (red arrow) decreased slightly towards postnatal stage P12. A similar pattern was observed for an anti-EPOP band at 40 kD (black arrow). It may therefore represent a somewhat variable degradation product of nuclear EPOP. In contrast, the 48 kD band of EPOP immunoreactivity (Fig.\u0026nbsp;4A; green arrow), was found to increase strongly in the postnatal stages. Thus, the form of EPOP corresponding to the cytoplasmic EPOP (compare Fig.\u0026nbsp;2A) became more abundant during development. Importantly and in line with the identified role of syndapin I as subcellular EPOP distribution modulator (Fig.\u0026nbsp;3), a similar increase of expression was observed for EPOP\u0026rsquo;s interaction partner syndapin I (Fig.\u0026nbsp;4A; Koch et al. 2011).\u003c/p\u003e \u003cp\u003eAlso in developing cultures of dissociated embryonic neuronal cultures, a comparable increase in expression of particularly the non-nuclear form of EPOP was observable (Fig.\u0026nbsp;4B). The additional anti-EPOP band at 40 kD showed a developmental decline and was almost undetectable at later postnatal stages (Fig.\u0026nbsp;4A) and in mature cultured neurons (Fig.\u0026nbsp;4B). The development-dependent course of expression of EPOP in the neural system showed great similarity to those of the three synaptically occurring proteins syndapin I, ProSAP1 and synapsin 1, which were analyzed in parallel (Fig.\u0026nbsp;4B). This could reflect some role of EPOP in dendritic spine and/or postsynapse formation. Immunocytochemical studies of endogenous EPOP in hippocampal neurons revealed that EPOP was initially concentrated in the nucleus, but in mature neurons more and more also showed abundance in the MAP2-positive dendrites, in which EPOP overlapped with the localization of syndapin I (Supplementary Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAs an EPOP presence in specifically dendritic spines would allow for a functional cooperation with syndapin I in spine and synapse formation (Schneider et al., 2014), we next addressed whether a portion of the extranuclear EPOP would colocalize with the postsynaptic marker homer1. In addition to the again clear nuclear, somatic and dendritic localization, immunostaining of mature neurons indeed showed colocalizations of endogenous EPOP with homer1 (Fig.\u0026nbsp;4C). Applying, anti-EPOP and anti-syndapin I antibodies together with antibodies against an additional component of the postsynaptic density (PSD), PSD-95, we furthermore observed that EPOP colocalized with its interaction partner syndapin I in PSD95-positive dendritic spines (Fig.\u0026nbsp;4D).\u003c/p\u003e\n\u003ch3\u003eEPOP is crucial for dendritic spines and for proper spine head organization\u003c/h3\u003e\n\u003cp\u003eIn order to be able to address whether EPOP may indeed be critical for dendritic spine formation and/or maintenance, we next established EPOP RNAi. Quantitative, fluorescence-based Western blot analyses demonstrated that the expression of EPOP was successfully and significantly suppressed by particularly EPOP RNAi-5 (Fig.\u0026nbsp;5A). Examinations of mature rat primary neuronal cultures that had been incubated with anisomycin, a protein biosynthesis inhibitor, for different periods of time revealed a rather rapid degradation of EPOP. Quantitative immunoblotting analyses showed that, compared to the DMSO-treated control cultures, a significant reduction in EPOP protein levels (around \u0026minus;\u0026thinsp;63%) was already detectable after 12 hours of anisomycin treatment. Since almost 75% of EPOP had already been degraded after 24 h (Fig.\u0026nbsp;5B,C), an incubation time of 48 h was considered as fully sufficient to significantly reduce the endogenous expression of EPOP by RNAi and to analyze the resulting EPOP loss-of-function phenotypes.\u003c/p\u003e \u003cp\u003eThe depletion of endogenous EPOP by EPOP-RNAi-2 in hippocampal neurons led to a reduction in the density of all dendritic spines by approximately 18% compared to neurons transfected with scrambled RNAi (Fig.\u0026nbsp;5D,E). This effect was mainly due to a reduction of the density of mushroom-shaped dendritic spines by more than 24% compared to the control (Fig.\u0026nbsp;5F).\u003c/p\u003e \u003cp\u003eIn addition, the mushroom-shaped dendritic spines of EPOP-RNAi-2 transfected neurons showed significantly narrower heads (Fig.\u0026nbsp;5G).\u003c/p\u003e \u003cp\u003eAlmost the same effects could be observed with EPOP-RNAi-5. In particular the spine head width EPOP loss-of-function phenotype was even more pronounced and thus had a higher significance compared the control scr-RNAi (Fig.\u0026nbsp;5F,G). The density of all dendritic spines resembled the results of RNAi-2 in absolute numbers relatively well when RNAi-5 was used but showed only a trend towards reduction (Fig.\u0026nbsp;5E). Analogously to EPOP-RNAi, neurons transfected with syndapin I-RNAi showed somewhat similar dendritic spine phenotypes when compared to EPOP loss-of-function, i.e. a significant reduction in the density of all dendritic spines, especially the mushroom-shaped dendritic spines, and a significant reduction in the head width of mushroom-shaped dendritic spines (Fig.\u0026nbsp;5D-G).\u003c/p\u003e \u003cp\u003eTo confirm the specificity of these phenotypes, we performed rescue experiments, in which an RNAi-resistant mutant of EPOP was co-expressed with RNAi-5. Quantitative immunoblot analyses demonstrated that \u0026ndash; in contrast to the significant depletion of WT rat EPOP upon coexpression of EPOP-RNAi-5 - coexpressed EPOP* could be detected with a protein level comparable to that under coexpression of scr-RNAi in quantitative anti-EPOP immunoblotting studies (Fig.\u0026nbsp;6A,B).\u003c/p\u003e \u003cp\u003eAn IRES system was subsequently used in order to ensure simultaneous expression of the respective RNAi, the reporter mCherryF and, if appropriate, EPOP*. EPOP-RNAi-5_IRES showed similar effects compared to EPOP-RNAi-5 (Fig.\u0026nbsp;6C-F; compare Fig.\u0026nbsp;5). The density of all dendritic spines was slightly but not significantly reduced after EPOP-RNAi-5_IRES. A slight increase in the overall dendritic spine density was noted when EPOP-RNAi-5_IRES_EPOP* was expressed, which was not significant compared to the control but when compared to EPOP-RNAi-5_IRES (Fig.\u0026nbsp;6D). Furthermore, a highly significant reduction in the density of mushroom-type dendritic spines by -24% compared to the control could be observed for RNAi-5. This phenotype was abolished by the expression of EPOP* (Fig.\u0026nbsp;6E). Also, the reduction in the head width of mushroom-shaped dendritic spines induced by EPOP depletion could be reversed by coexpressing EPOP* and no significant difference from the control were observed anymore (Fig.\u0026nbsp;6F).\u003c/p\u003e \u003cp\u003eOverall, the experiments showed that the dendritic spine phenotypes induced by EPOP depletion with RNAi-5 were rescued by reexpressing EPOP*. This proved the specificity of the identified EPOP loss-of-function phenotypes in dendritic spine formation and organization.\u003c/p\u003e\n\u003ch3\u003eDepletion of EPOP causes a reduction of homer1 in dendritic spines\u003c/h3\u003e\n\u003cp\u003eThe results obtained so far raised the important question whether and to what extent not only the morphology of dendritic spines was affected by the depletion of EPOP, but also its PSD, and thus a\u003c/p\u003e \u003cp\u003epossible functional limitation was present. The analysis of the postsynaptic protein homer1 showed a strong impairment in dendritic spines after EPOP RNAi (Fig.\u0026nbsp;7). EPOP-RNAi-5 resulted in a 48% reduction in homer1 VOI-Rs in dendritic spines compared to control (Fig.\u0026nbsp;7B). Only 41% of the mushroom-type dendritic spines still showed a homer1 VOI-R after EPOP-RNAi-5, compared to 73% in the control (scr-RNAi; relative: -44%, Fig.\u0026nbsp;7C). The homer1-positive VOI-Rs that still were detected in dendritic spines furthermore showed a reduced intensity of homer1 and were significantly smaller (Fig.\u0026nbsp;7E, F). Consequently, depletion of EPOP with RNAi-5 resulted in a drastic reduction of the postsynaptic scaffold component homer1 in dendritic spines.\u003c/p\u003e \u003cp\u003eInterestingly, syndapin I RNAi resulted in a similarly clear impairment of the homer1 content in dendritic spines as EPOP-RNAi-5 (Fig.\u0026nbsp;7). The general occurrence in dendritic spines and explicitly in mushroom-shaped dendritic spines, was massively reduced (Fig.\u0026nbsp;7B,C). In addition, the intensity of homer1 in the remaining VOI-Rs (Fig.\u0026nbsp;7D) as well as their volume (Fig.\u0026nbsp;7E) was significantly reduced. Syndapin I loss-of-function phenotypes in dendritic spine formation and organization thus mirrored those of its interaction partner EPOP in neurons and thereby highlighted their also functional relationship in postsynapses. Both EPOP and syndapin I depletion thus caused a massive reduction in homer1 in mature dendritic spines. Taking into account the reduction of the density of especially mushroom-shaped dendritic spines upon EPOP or syndapin I RNAi (compare Fig.\u0026nbsp;5), the postsynaptic proteins along the dendrite were even more strongly reduced.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eEPOP is a mammalian-specific protein that previously had been defined due to its association with PRC2 (Zhang et al., 2011; De Cegli et al., 2013; Smits et al., 2013; Alekseyenko et al., 2014). Consistent with the PRC2 providing a crucial epigenetic mechanism for regulating embryonic development, EPOPs localization has been reported as nuclear in embryonic stem cells (Liefke \u0026amp; Shi 2015; Behringer et al., 2016; Liefke et al., 2016; Healy et al., 2019). We here report a moonlighting function for EPOP in mature neurons outside the nuclear compartment together with its herein identified binding partner syndapin I. Our data clearly demonstrate that in mature neurons \u0026ndash; both in dissociated cultures as well as in intact tissue in various brain regions \u0026ndash; EPOP in addition adopts a somatodendritic localization. Biochemical fractionation data confirmed that EPOP is present in both nuclear and extranuclear compartments. The colocalization with MAP2 furthermore demonstrated that EPOP is preferentially expressed in neurons, whereas no prominent overlap with glial cells identified both in cultures and in brain section by anti-GFAP labeling could be observed. It may well be that these observations in mature neurons represent a cell biological specialty, as neurons are both highly differentiated cells and postmitotic, yet they do express relatively high amounts of EPOP. It therefore seems that PRC2 functions in embryonic stem cells and extranuclear functions of the accessory PRC2 component EPOP represent completely distinct functional aspects.\u003c/p\u003e \u003cp\u003eThe extranuclear role of EPOP in mature neurons appears to involve EPOP interactions with syndapin I. This finding is based on several experimental results. The identification of EPOP/syndapin I complexes by Y2H screening was underscored by coprecipitation of GFP-EPOP from cell lysates by recombinant syndapin I. EPOP associated with the SH3 domain of syndapin I, which was both critical but also sufficient for the interaction. Given the fact that EPOP is rich of prolines and also harbors a variety of PxxP-motifs, it appears likely that the identified interaction is based on a canonical SH3/PxxP association mode. EPOP indeed directly interacted with syndapin I, as proven by in vitro reconstitution with purified components. The in vivo relevance of the interaction was highlighted by heterologous coimmunoprecipitations showing effective association of syndapin I with EPOP. The fact that EPOP was able to coprecipitate endogenous syndapin I from brain extracts furthermore demonstrated the importance of the interaction in the adult brain. Importantly, our mitochondrial recruitment experiments ruled out that the observed syndapin I/EPOP interaction in the biochemical experiments represent putatively possible post-solubilization artifacts due to disruption of nuclear integrity and release of nuclear components. Instead, these experiments visually demonstrated specific complex formation in fully intact cells. The in vivo-recruitment assays also revealed that, if high-affinity extranuclear docking sites for EPOP are provided, the equilibrium of extranuclear and nuclear EPOP can readily be shifted. The mitochondrial recruitment assays are thereby very well in line with our observations of shifts in cytoplasmic/nuclear distribution observed both upon syndapin I overexpression and conversely also upon deficiency of syndapin I in neurons.\u003c/p\u003e \u003cp\u003eSeveral aspects make syndapin I particularly suitable as extranuclear EPOP anchor, it\u0026rsquo;s efficient, SH3-based association with EPOP, the fact that syndapin I is binding to the plasma membrane via its N-terminal F-BAR domain (Itoh et al., 2005; Dharmalingam et al., 2009) and the special property that syndapin I acts as a dimer (Kessels \u0026amp; Qualmann, 2006), i.e. a syndapin I dimer has the potential to bind to EPOP and to simultaneously associate with further cellular components via an SH3 domain interaction. Remarkably, neither in previous studies nor in comparative biochemical and immunocytochemical analyses in parallel to EPOP, syndapin I has ever been detected in nuclear compartments. It thus seems that syndapin I provides extranuclear anchor points for EPOP and that EPOP undergoes nucleocytoplasmic shuttling. In line, in silico analyses suggest both nuclear localization signals (NLS) and a nuclear export signal (NES) to be present in EPOP (our unpublished data). Comparisons of the immunoreactivity of endogenous EPOP in \u003cem\u003ewildtype\u003c/em\u003e (WT) and \u003cem\u003esyndapin I\u003c/em\u003e KO neurons revealed that syndapin I is critical for keeping a portion of EPOP extranuclear. Together, our experiments demonstrated that syndapin I is not only capable but \u0026ndash; under physiological conditions \u0026ndash; also required for proper intracellular distribution of EPOP in mature neurons.\u003c/p\u003e \u003cp\u003eInterestingly, EPOP was detected to overlap with the dendritic marker MAP2 but not with the axonal marker SMI312 in both mouse brain sections and in primary hippocampal rat cultures. EPOP thus seems to have a high preference for the dendritic/postsynaptic compartment. This observation distinguishes EPOP from its binding partner syndapin I, which also functions in axonal morphogenesis (Dharmalingam et al., 2009) and in presynapses (Koch et al., 2011).\u003c/p\u003e \u003cp\u003eComparable to its interaction partner syndapin I and further synaptic proteins, such as synapsin I and ProSAP1/Shank2, EPOP expression increases upon neuronal development and showed highest levels in fully developed and differentiated neurons. Our observations at the protein level are in line with in situ hybridizations in mouse brains, that also highlighted a prominent presence of EPOP mRNA in cortical and hippocampal neurons not only at the embryonic but also postnatally, at a juvenile age (P21) (De Cegli et al. 2013). Upon in vitro differentiation of mouse embryonic stem cells, in contrast, a decline of EPOP was observed (De Cegli et al., 2013; Liefke et al., 2016). This apparent difference to our result in differentiating neurons supports a distinct role of EPOP in differentiated neurons independent form its earlier reported functions in PRC2 and ElonginB/C regulation in the nucleus. In line with such a distinct role, at least in neurons, the main proteins of the canonical PRC2, such as enhancer of zeste homolog (EZH) 2 or suppressor of zeste 12, exhibited a rapid decrease in expression during neuronal development (Henriquez et al., 2013). EPOP has been precipitated with both EZH2 and EZH1, although EZH1 and EZH2 can be excluded as simultaneous components of a PRC2 (Zhang et al., 2011; Alekseyenko et al., 2014; Margueron et al., 2008; Shen et al., 2008). Interestingly, EZH2 induces the expression of PSD95, a postsynaptic scaffold protein (Henriquez et al., 2013). Yet, only for EZH1, a non-canonical major protein of PRC2, a more long-lasting neuronal expression has been described (Henriquez et al., 2013). EPOP as well as its interaction partner syndapin I were observed to rise in expression towards neuronal maturation and towards the formation of neuronal networks by synapse formations.\u003c/p\u003e \u003cp\u003eIn the dendritic compartment of mature neurons, EPOP was observed to localize to dendritic spines, where its immunoreactivity overlapped with the PSD proteins PSD95 and homer1 and with its binding partner syndapin I. The dendritic spine offers several specific aspects bringing about elevated levels of physically attached syndapin I. Syndapin I interacts with the postsynaptic scaffold protein ProSAP1/Shank2 (Schneider et al., 2014) and syndapin I\u0026rsquo;s membrane anchoring shows a preference for dendritic spines and their different membrane curvatures (Schneider et al., 2014). Loss-of-function studies in primary hippocampal neurons demonstrated that EPOP was particularly important for mushroom-type spines. RNAi-mediated knockdown of EPOP significantly reduced the density of mushroom spines, a phenotype consistently observed for two different RNAi tools. Rescue experiments with an RNAi-resistant version of EPOP showed the specificity of the loss-of-function analyses.\u003c/p\u003e \u003cp\u003eInterestingly, also knockdown of the EPOP interaction partner syndapin I resulted in a decreased mushroom spine density in DIV18 neurons. In previous studies, similar defects had been observed in less mature dissociated hippocampal neurons (DIV14) (Schneider et al., 2014). EPOP loss-of-function \u0026ndash; and consistently likewise syndapin I loss-of-function \u0026ndash; did not only result in fewer mushroom spines, but those present had significantly decreased spine head sizes.\u003c/p\u003e \u003cp\u003eThe dendritic spine heads harbor the postsynaptic scaffold and signal integration and organization machinery. Impairments in PSD organization upon syndapin I-deficiency had been described previously. Syndapin I deficient neurons showed a significant reduction of PSD95 points along the dendrites of younger, DIV14 neurons (Schneider et al. 2014). Since previous studies had reported influences of PRC2 components on PSD95 mRNA expression levels through transcriptional control (Henriquez et al., 2013), we did not address putative phenocopying defects upon EPOP deficiency by analyzing PSD-95. Instead, we analyzed the major synaptic scaffold protein homer1 as an indicator for proper PSD organization. Analyses of homer1 and the morphology of spines clearly showed that the depletion of endogenous EPOP, analogous to a depletion of syndapin I, resulted in a massive impairment of mature spines and their postsynaptic organization. The mushroom spines still remaining upon EPOP RNAi or syndapin I RNAi exhibited additional defects in the composition and organization of the postsynaptic density, they contained significantly lesser or even no detectable levels of the major synaptic scaffold protein homer1. Both the volume and the mean intensity of homer1-positive structures in mushroom spine heads were highly decreased. The fact that upon loss of EPOP or syndapin I, the integrity of dendritic spines and of postsynapses was severely impaired indicated that both proteins not only play an important role in the formation of spines but are also responsible for the maintenance and stabilization of mature spines. The broad correspondence of the phenotypes on spines and postsynapses induced by syndapin I and EPOP RNAi and their identified direct interaction strongly suggests that EPOP and syndapin I jointly stabilize mature spines as well as their postsynapses.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003ePlasmids\u003c/h2\u003e \u003cp\u003eFull-length mouse (\u003cem\u003emus musculus\u003c/em\u003e) EPOP (NCBI NP_780541.1) constructs were generated by performing PCR reactions on a mouse cDNA library (\u003cem\u003eBD\u003c/em\u003e Biosciences λTriplEx\u0026trade;) and subsequently cloned into pCMV-Tag2B (Stratagene), pEGFP-C2 (Clontech) and pIRES2-eGFP (BD Bioscience/Clontech). For mitochondrially targeted EPOP constructs, EPOP was subcloned into Mito-Flag-mCherry-pCMV-Tag2B (Mito-mCherry), a mitochondrial targeting vector described previously (Hou et al., 2015).\u003c/p\u003e \u003cp\u003eFurther EPOP constructs were obtained by subcloning inserts from Y2H clones into pGEX-4T2, pEGFP-CW3 and pET32c (Novagen). GFP-EPOP\u003csup\u003e1\u0026ndash;169\u003c/sup\u003e was amplified by PCR using mouse full-length EPOP as template with the primers 5\u0026rsquo;-cggaattcatggagactctgtgtcctcct-3\u0026rsquo; and 5\u0026rsquo;-cgcgtcgacctaactgctagctgcatcaagacc-3\u0026rsquo; and subcloned into pEGFP-C2. A corresponding GST-fusion was obtained by subcloning into pGEX-5X-1.\u003c/p\u003e \u003cp\u003eRat (\u003cem\u003erattus norvegicus\u003c/em\u003e) EPOP (NCBI NP_001103097.1 with G117V und R199P sequence variants corresponding to the sequence variants deposited for M0RD31_RAT in UniProt variant viewer) was amplified from adult rat cortex cDNA with the primers 5\u0026rsquo;-ccggaattcatggagactctgtgtcc-3\u0026rsquo; and 5\u0026lsquo;-acgcgtcgactcagagttcttccaag-3\u0026lsquo; and subcloned into pEGFP-C2 and pIRES2-eGFP (BD Bioscience/Clontech).\u003c/p\u003e \u003cp\u003eRNAi constructs directed against EPOP were generated according to the methods described previously (Ahuja et al., 2007). In brief, phosphorylated primers for EPOP-RNAi-2 (target sequence rat nt 761\u0026ndash;785), 5\u0026rsquo;- gatccaaacttggagtgtccagggcgaaccttgatatccgggttcgccctggacactccaagttttttttta-3\u0026rsquo; and 5\u0026rsquo;- agcttaaaaaaaaacttggagtgtccagggcgaacccggatatcaaggttcgccctggacactccaagtttg-3\u0026lsquo; and EPOP-RNAi-5 (target sequence rat nt 741\u0026ndash;761) 5\u0026lsquo;-GATCCACTTTGGTGTTACGCGAAAGGACTCGAGAcctttcgcgtaacaccaaagtTTTTTA-3\u0026rsquo; and 5\u0026rsquo;- AGCTTAAAAAACTTTGGTGTTACGCGAAAGGTCTCGAGTcctttcgcgtaacaccaaagtG-3\u0026rsquo; were annealed. Subsequently, the products were subcloned into pRNAT-H1.1 coexpressing farnesylated mCherry (mCherry-F) (pRNAT-H1.1/mCherryF) or eGFP (pRNAT-H1.1/GFP). Corresponding RNAi vectors expressing a scrambled RNAi sequence (scr-RNAi) served as controls (Nolze et al., 2013). For an analysis of knockdown efficiency of coexpressed rat EPOP in HEK293 cells via Western blot analyses, rat EPOP was subcloned from pIRES2-eGFP together with the IRES sequence using AfeI/NotI into pRNAT-H1.1/GFP expressing scr-RNAi, EPOP-RNAi-2 and EPOP-RNAi-5, respectively, generating scr-RNAi_GFP_IRES_EPOP, EPOP-RNAi-2_GFP_IRES_EPOP, and EPOP-RNAi-5_GFP_IRES_EPOP, respectively.\u003c/p\u003e \u003cp\u003eFor rescue experiments, the IRES sequence from pIRES2-eGFP was amplified using the primers 5\u0026lsquo;-atagagctcgatccgcccctctccctccccccc-3\u0026lsquo; and 5\u0026lsquo;-atacccgggttaattaattaactagttgtggccatattatcatcgtgttt-3\u0026lsquo; and cloned into scr-RNAi and EPOP-RNAi-5 in pRNAT-H1.1/mCherryF, respectively, generating scr-RNAi_IRES and EPOP-RNAi-5_IRES, respectively. RNAi-resistant mouse EPOP containing silent mutations (EPOP*) was generated by mutagenesis PCR using the mutagenesis primers 5\u0026rsquo;-aaccgcctgatccgccgaagtaagttatggtgctatgccaagggcttcgccctggacact-3\u0026rsquo; and 5\u0026rsquo;-agtgtccagggcgaagcccttggcatagcaccataacttacttcggcggatcaggcggtt-3\u0026rsquo; and subcloned into pIRES2-eGFP and subsequently together with the IRES sequence using AfeI/NotI into pRNAT-H1.1/GFP expressing scr-RNAi and EPOP-RNAi-5, respectively, generating scr-RNAi_GFP_IRES-EPOP* and EPOP-RNAi-5_GFP_IRES-EPOP* respectively, and into EPOP-RNAi-5_IRES generating EPOP-RNAi-5_IRES_EPOP*.\u003c/p\u003e \u003cp\u003eBD-syndapin I and BD-syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e cloned in the pGBTK7 vector have been described in Braun et al. (2005). GST-syndapin (Izadi et al., 2021), GST-syndapin I\u003csup\u003eSH3\u003c/sup\u003e (Braun et al., 2005) and GST-syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e (Qualmann et al., 1999) were described previously. GST-syndapin I\u003csup\u003eBAR\u003c/sup\u003e (aa1-305) was amplified by PCR ad subcloned into pGEX-5X-1. GFP-syndapin I was described in Kessels \u0026amp; Qualmann (2006). Xpress-tagged syndapin I and syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e were described in Qualmann \u0026amp; Kelly (2000). RNAi constructs against syndapin I have been described previously (Dharmalingam et al., 2009).\u003c/p\u003e \u003cp\u003eAll PCR-based constructs were verified by sequencing.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eProteins\u003c/h2\u003e \u003cp\u003eGST- and TrxHis-tagged fusion proteins were purified from E. coli as described previously (Qualmann \u0026amp; Kelly, 2000; Schwintzer et al., 2011).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eAntibodies\u003c/h2\u003e \u003cp\u003eAntisera against EPOP were raised in rabbit (Pineda Antik\u0026ouml;rper-Service, Germany) using a purified GST-fusion protein of amino acids 52\u0026ndash;369 of mouse EPOP as antigen. Using recombinant TrxHis-EPOP\u003csup\u003e52\u0026ndash;369\u003c/sup\u003e immobilized on a CNBr sepharose 4b (Amersham) column and acidic elution, affinity-purified polyclonal anti-EPOP antibodies were obtained.\u003c/p\u003e \u003cp\u003ePolyclonal rabbit and guinea pig antibodies against GST-syndapin I as well as anti-GST and anti-TrxHis antibodies were purified from antisera as described previously (Qualmann et al., 1999; Braun et al., 2005; Izadi et al., 2021).\u003c/p\u003e \u003cp\u003eMonoclonal mouse anti-GFP antibodies (JL-8) were from Clontech/Takara (632381, Takara). Monoclonal anti-Xpress antibodies (R910-25) were from Invitrogen. Polyclonal rabbit anti-GFP antibodies (ab290) were from Abcam.\u003c/p\u003e \u003cp\u003eMonoclonal mouse (M2; F3165) and polyclonal rabbit anti-Flag (F7425) antibodies, monoclonal mouse anti-MAP2 (HM-2; M4403), monoclonal mouse anti-\u0026szlig;-tubulin (clone TUB 2.1) anti monoclonal mouse anti-GFAP (G3893) antibodies as well as polyclonal rabbit anti-ProSAP1 (anti-SHNAK2; HPA008174) antibodies were from Sigma-Aldrich\u0026reg; Co. LLC. Guinea pig anti-MAP2 and anti-homer1 antisera and mouse monoclonal anti-synapsin1 antibodies (clone 46.1) from Synaptic Systems. Mouse monoclonal anti-fibrillarin antibodies (clone 38F3). Mouse monoclonal anti-PSD95 (clone 6G6-1C9) antibodies and rabbit polyclonal anti-laminB1 antibodies were from Abcam. Goat polyclonal anti-GAPDH antibodies were from Santa Cruz Biotechnology. Mouse monoclonal anti-SMI312 (837904) antibodies were from Covance/Biozol.\u003c/p\u003e \u003cp\u003eNormal rabbit IgG (10500C) was from Thermo Fischer Scientific.\u003c/p\u003e \u003cp\u003eSecondary antibodies included Alexa Fluor488-labeled goat anti-mouse (A-10680), donkey anti-mouse (A-21202), Alexa Flour586-labeled donkey anti-mouse (A10037), donkey anti-rabbit (A10042), Alexa Flour647-labeled donkey anti-mouse (A-31571) and donkey anti-rabbit (A-31573) antibodies (Thermo Fisher Scientific) and Alexa Flour647-labeled donkey anti-guinea pig antibodies (06-605-148, Dianova).\u003c/p\u003e \u003cp\u003eFurther secondary antibodies used included AlexaFluor680-labeled goat anti-rabbit and anti-mouse antibodies and AlexaFluor680-labeled donkey anti-goat antibodies (A-32734, A-21058; A-21084; Thermo Fisher Scientific); DyLight800-conjugated goat anti-rabbit and anti-mouse antibodies (SA5-35571 and SA5-35521; Thermo Fisher Scientific), donkey anti-guinea pig antibodies coupled to IRDye800, (926-32411; LI-COR Bioscience) and peroxidase-conjugated goat anti-rabbit antibodies (111-035-045; Dianova).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eIsolation of RNA and reverse transcription\u003c/h2\u003e \u003cp\u003eThe RNA isolation from adult rat cortex was carried out according to Haag et al. (2012). For reverse transcription the RevertAid H minus First Strand cDNA Synthesis Kit was used according to the manufacturer's instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eIn vitro reconstitutions of direct protein\u0026ndash;protein interactions\u003c/h2\u003e \u003cp\u003eDirect protein\u0026ndash;protein interactions were demonstrated by coprecipitations with combinations of recombinant TrxHis- and GST-tagged fusion proteins purified from \u003cem\u003eE. coli\u003c/em\u003e as described in Izadi et al. (2018). In brief, complex formation of TrxHis-EPOP\u003csup\u003e52\u0026ndash;369\u003c/sup\u003e and TrxHis-EPOP\u003csup\u003e222\u0026ndash;369\u003c/sup\u003e with GST-syndapin I\u003csup\u003eSH3\u003c/sup\u003e was demonstrated in 10 mM HEPES, pH 7.4, 300 mM NaCl, 0.1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, and 1% (v/v) Triton X-100 supplemented with EDTA-free protease inhibitor cocktail. Eluted proteins were analyzed by SDS-PAGE, transferred to polyvinylidene fluoride membranes and then analyzed by immunodetection with anti-TrxHis and anti-GST antibodies by using a Licor Odyssey System.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003ePreparation of brain extracts\u003c/h2\u003e \u003cp\u003eRat brains were homogenized with an ultraturrax in HEPES buffer (10 mM HEPES, 1 mM EGTA, 0.1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 0.15 M NaCl, 1x protease inhibitor without EDTA, pH 7.5; 3 ml per g tissue) for several times for 15\u0026ndash;20 seconds at 20000 rpm under ice cooling. The supernatant of a subsequent centrifugation at 150000 x g and 4\u0026deg;C for 45 minutes was obtained as rat brain extract.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePreparation and fractionation of mouse tissue homogenates\u003c/h2\u003e \u003cp\u003eTissue homogenates in RIPA buffer were prepared as described previously (Hou et al., 2018).\u003c/p\u003e \u003cp\u003eFor fractionation of mouse cortex homogenates, mouse cortices were prepared and homogenized by 12 strokes with the Potter S homogenizer at 900 min\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e in 10 \u0026micro;l homogenization buffer (5 mM HEPES pH7.4, 1 x PIC EDTA-free, 0.32 M sucrose, 1 mM EDTA) per mg wet weight. The homogenate was centrifuged for 10 min at 1000 x g and 4\u0026deg;C. Then, 100 \u0026micro;l as supernatant were removed. The sediment was homogenized in the same volume as before. The mixture was then centrifuged for 10 min at 1000 x g and 4\u0026deg;C. The pellet was resuspended in 1 ml PBS / PIC. The fractions were analyzed by SDS-PAGE and Western blot.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLysis of primary cortical rat neurons\u003c/h2\u003e \u003cp\u003eCultured neurons were washed once with cold PBS, detached using a cell scraper and taken up in 1 ml cold PBS. The cells were centrifuged for 5 min at 1000 x g and 4\u0026deg; C and the pellet was resuspended in 400 \u0026micro;l lysis buffer (10 mM HEPES, 1 mM EGTA, 0,1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 10 mM NaCl, 1% (v/v) Triton\u0026reg; X-100, 1 x PIC complete; pH 7.4). After 5 ultrasound pulses (1 s each) and 20 min incubation at 4 \u0026deg; C on a rotary wheel, the samples were centrifuged at 4\u0026deg;C for 10 min at 20800 x g. Subsequently, 400 \u0026micro;l of the supernatant were mixed with 1.6 ml -20\u0026deg;C cold acetone. The sample was incubated overnight at -20\u0026deg;C. The precipitated proteins were sedimented (30 min at 12000 x g and 4\u0026deg;C) and the pellet was washed with 2 ml of -20\u0026deg; C cold 80% (v/v) ethanol. After drying for two minutes, the pellet was resuspended in 100 \u0026micro;l 1 x SDS sample buffer. In order to facilitate the resuspension, 3 ultrasound pulses of 1 s were applied. Finally, the samples were incubated for 5 min at 95\u0026deg; C and subsequently analyzed by immunoblotting.\u003c/p\u003e \u003cp\u003eFor the analysis of the time course of degradation of endogenous EPOP, cortical neurons (DIV17) were treated for different time intervals (2\u0026ndash;24 h) with the protein synthesis inhibitor anisomycin (7.5 \u0026micro;M final). Cell lysates were prepared as described above. Lysates of neuronal cultures incubated with DMSO (0.2% (v/v) final) were used as solvent control.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eYeast two-hybrid analyses\u003c/h2\u003e \u003cp\u003eYeast two-hybrid screening was carried out using the GAL4-based Matchmaker yeast two-hybrid system 3 (BD Biosciences Clontech) with full-length syndapin I as a bait and a pretransformed mouse brain library (BD Biosciences Clontech) as described in Braun et al. (2005). Subsequently, prey plasmids were isolated, retransformed into yeast, and mated with yeast strains transformed with BD-syndapin I, BD-syndapin I\u003csup\u003eΔSH3\u003c/sup\u003e, and with the pGBTK7 vector encoding for the BD domain alone, respectively. The obtained diploids were subsequently assayed for the activation of reporter genes according to Kessels \u0026amp; Qualmann (2002).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eAnimals\u003c/h2\u003e \u003cp\u003eRats (Crl:WI; Charles River) were used for primary cell culture preparations and biochemical analyses from brain tissue. The generation and analysis of \u003cem\u003esyndapin I\u003c/em\u003e KO mice has been described previously (Koch et al., 2011; Koch et al., 2020).\u003c/p\u003e \u003cp\u003e All animal breeding was in strict compliance with the EU directives 86/609/EWG and 2007/526/EG guidelines for animal experiments and all related procedures were approved by the local government (Th\u0026uuml;ringer Landesamt, Bad Langensalza, Germany, and Landesverwaltungsamt, breeding allowance 02-0571/14 and UKJ-17-021).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eCell culture, transfection and immunostaining\u003c/h2\u003e \u003cp\u003eHEK293 and COS-7 cells were maintained in 10 ml DMEM containing 2 mM L-glutamine, 10% (v/v) fetal bovine serum and penicillin/streptomycin (Invitrogen) at 37\u0026deg;C, 90% humidity and 5% CO\u003csub\u003e2\u003c/sub\u003e. Cells for immunofluorescence-based analyzes were processed on coverslips coated with poly-D-lysine. HEK293 and COS-7 cells were transfected with plasmid DNA using TurboFect (Thermo Fisher Scientific) according to manufacturer\u0026acute;s instructions.\u003c/p\u003e \u003cp\u003eImmunolabeling of cultured cells was essentially as described in Hou et al. (2018). Mitochondria of living cells were visualized by incubating the cells with a final concentration of 0.2 \u0026micro;M MitoTracker Deep Red 633 for 1 h prior to fixation.\u003c/p\u003e \u003cp\u003ePreparation of HEK293 cell lysates for coprecipitation and coimmunoprecipitation analyses was performed as described in Qualmann \u0026amp; Kessels (2006), Schwintzer et al. (2011) and Izadi et al. (2018).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eCoprecipitation analyses\u003c/h2\u003e \u003cp\u003eCoprecipitation analyses were performed with HEK293 cell lysates containing overexpressed GFP-fusion proteins or with mouse brain extracts and GST-tagged proteins immobilized to glutathione resin (Genscript Corp.) as described before (Qualmann et al., 1999; Schwintzer et al., 2011).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eCoimmunoprecipitation\u003c/h2\u003e \u003cp\u003e \u003cstrong\u003eCoi\u003c/strong\u003e \u003cp\u003emmunoprecipitation analyses of Flag-EPOP and GFP-syndapin I were performed from lysates from transfected HEK293 cells as described previously in Schwintzer et al. (2011). A lysis buffer containing 120 mM NaCl was used.\u003c/p\u003e \u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003ePreparation and cultivation of primary neuronal cultures\u003c/h2\u003e \u003cp\u003ePreparation of primary rat neurons and their cultivation was carried out according to the protocols of Banker and Cowan (1977) and Brewer et al. (1993). The protocol used was as described in Schneider et al. (2014). Primary hippocampal mouse neurons (DIV17) of wild-type and syndapin I-deficient animals were prepared as described in Koch et al. (2020). Rat cortical neurons were prepared and cultured as described in Wolf et al. (2019).\u003c/p\u003e \u003cp\u003eImmunolabeling of cultured cells was essentially as described in Schwintzer et al. (2011) and Schneider et al. (2014).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eTransient transfection of rat neurons\u003c/h2\u003e \u003cp\u003eAfter 16 days in culture, calcium phosphate transfections were performed as described previously (Schneider et al., 2014). For some experiments, the DIV16 neurons were alternatively transiently transfected using Lipofectamine\u0026reg; 2000 as described before (Schwintzer et al., 2011; Qualmann et al., 2004).\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eImmunolabeling of mouse brain sections\u003c/h2\u003e \u003cp\u003eMouse brain sections were prepared essentially as described in Haag et al. (2012) and Schneider et al. (2014). Immunolabeling of brain sections was performed according to Schneider et al. (2014) except that PBS instead of PB was used for all solutions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eLight microscopy\u003c/h2\u003e \u003cp\u003eThe images of cells and brain sections were taken using a Zeiss AxioObserver.Z1 microscope. In order to achieve pseudo confocality, Z stacks were recorded with an ApoTome, at distances of 0.24 \u0026micro;m (63x/1.4 objective), 0.3 \u0026micro;m (40x/1.3 objective) and 1 \u0026micro;m (20x/0.5 objective), respectively. Images were recorded digitally and either AxioVision or Zen software was used for image processing. For the overview picture of a mouse brain, the Tiles mode of the Zen software was applied with an overlap of 10%; all tiles in the 3 Z planes were recorded at a distance of 10 \u0026micro;m. All single images were stitched together by automatic stitching using Zen software. The overview is shown as maximum intensity projection (MIP).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eInvestigation of the subcellular distribution of overexpressed EPOP in COS-7 cells\u003c/h2\u003e \u003cp\u003eCOS-7 cells were cotransfected with a plasmid encoding for GFP-EPOP in combination with mCherryF or Xpress-syndapin I. Anti-EPOP intensity measurements were based on anti-EPOP immunostaining using Fiji. As a marker for nuclear nuclei acids, DAPI was applied. The entire cell was marked by the reporters mCherryF or syndapin I, respectively. The mean intensity for the cytoplasm was calculated from the mean intensities and the areas of both regions (whole cell, nucleus) by subtraction.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eDistribution analysis of endogenous EPOP in wild-type and syndapin I-deficient neurons\u003c/h2\u003e \u003cp\u003ePrimary hippocampal mouse neurons (DIV17) of wild-type and syndapin I-deficient animals prepared as described in Koch et al. (2020) were fixed and stained for EPOP, MAP2 and fibrillarin as well as with DAPI and subjected to pseudoconfocal imaging (ApoTome). The evaluations of the Z stacks were carried out blindly. Using Imaris software, fibrillarin and MAP2 immunoreactivity as well as DAPI signals were used to create so-called surfaces of volume-of-interest reconstructions (VOI-Rs). These surface reconstructions represented either the nucleoli (fibrillarin), the nucleus (DAPI) or the (somatodendritic) cytoplasm (MAP2). The mean intensity of endogenous anti-EPOP immunosignals was then measured in these VOI-Rs. Taking into account the VOI-R volume, the mean intensities of the nucleoplasm and of the entire cells were calculated. Subsequently, the ratio of the mean intensities of two compartments was calculated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eMorphological analysis of spines\u003c/h2\u003e \u003cp\u003eSpines were analyzed using rat hippocampal neurons that were transfected at DIV16 and further cultured for 48 h. The neurons were then immunostained and imaged as described above. One dendrite section having a minimum length of 100 \u0026micro;m was evaluated per cell. The investigations were carried out blindly. Imaris software was used to reconstruct and define all spines of the dendritic segment. The diameter of both the dendrite and the spine was calculated using the shortest distance algorithm. The parameters for the detection of the spines were selected as follows: minimum diameter 0.1 \u0026micro;m and maximum length 5 \u0026micro;m. If necessary, manual corrections were made to trace individual spines correctly in 3D. The spines were then grouped using the spine classifier according to their subtypes. The following settings were made for this: Base of the spines: proximal 25% of the spines; head of the spine: distal 25% of the spine; stubby spines: spines length\u0026thinsp;\u0026le;\u0026thinsp;0.75 \u0026micro;m; mushroom-shaped spines: spines length\u0026thinsp;\u0026gt;\u0026thinsp;0.75 \u0026micro;m and maximum head width\u0026thinsp;\u0026gt;\u0026thinsp;0.5 \u0026micro;m; thin spines: spines length\u0026thinsp;\u0026gt;\u0026thinsp;0.75 \u0026micro;m and maximum head width\u0026thinsp;\u0026le;\u0026thinsp;0.5 \u0026micro;m and maximum head width\u0026thinsp;\u0026gt;\u0026thinsp;0.3 \u0026micro;m and average neck width\u0026thinsp;\u0026lt;\u0026thinsp;maximum head width; filopodia-like spines: all other spines detected. All parameters were normalized to the values for the respective control in the blinded analyses.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eInvestigation of postsynaptic proteins in spines\u003c/h3\u003e\n\u003cp\u003eTo investigate the expression of homer1 in the spines, a created Imaris reconstruction (\u003cem\u003efilament\u003c/em\u003e) as well as only the dendrite of the filament were each converted into individual fluorescence channels. A VOI-R of the dendrite was then created. This was used for masking (voxel within the VOI-R to 0) of the filament channel, so that a new fluorescence channel, which contained only spines, could be generated. A colocalization channel was then formed from the spines channel and the homer1 channel. On the basis of this colocalization channel, VOI-Rs for homer1 were generated, which reflected the signal on postsynaptically localized protein above a manually defined, constant threshold value.\u003c/p\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003eStatistics\u003c/h2\u003e \u003cp\u003eAll statistical evaluations carried out in this work were carried out in GraphPad Prism. Whenever applicable (n number restriction), the Shapiro-Wilk test was used to first check the values for normal distribution. Otherwise, the Kolmogorov\u0026ndash;Smirnov normality distribution test was used. If the values were distributed normally, a t-test (2 conditions to compare) and a one-way ANOVA (\u0026ge;\u0026thinsp;3 conditions to compare), respectively, were used. For not normally distributed data, a Mann-Whitney U test (2 samples) or a Kruskal-Wallis test (\u0026ge;\u0026thinsp;3 samples) was carried out. Both multiple comparisons were followed by posttests, Bonferroni's Multiple Comparison Test for one-way ANOVA evaluations and a Dunn's Multiple Comparison Test for Kruskal-Wallis tests. Two-way ANOVA examinations were carried out for variance analyses with two factors.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank B. Schade A. Kreusch, M. Öhler and C. Scharf for excellent technical assistance and D. Koch, E. Wünsche and A. Inciute for experimental support. This work was supported by DFG grants QU116/5-2 and 11-1 to BQ and KE685/7-1 to MMK.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJ.D. conducted the vast majority of experiments shown. R.D. made initial observations and also conducted experiments. N.K. generated primary mouse cultures.J.D, M.M.K, and B.Q. interpreted and visualized data. J.D. cowrote an initial draft of the manuscript. B.Q. and M.M.K conceived the project, provided scientific supervision and funding and wrote the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of interests:\u0026nbsp;\u003c/strong\u003eThe authors declare no competing financial interests.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAguilar, R., Bustos, F. J., Saez, M., Rojas, A., Allende, M. L., van Wijnen, A. J., van Zundert, B. \u0026amp; Montecino, M. Polycomb PRC2 complex mediates epigenetic silencing of a critical osteogenic master regulator in the hippocampus. \u003cem\u003eBiochim. Biophys. Acta \u003c/em\u003e\u003cstrong\u003e1859\u003c/strong\u003e, 1043\u0026ndash;1055 (2016).\u003c/li\u003e\n\u003cli\u003eAhuja, R., Pinyol, R., Reichenbach, N., Custer, L., Klingensmith, J., Kessels, M.M. \u0026amp; Qualmann, B. Cordon-bleu is an actin nucleation factor and controls neuronal morphology. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e131\u003c/strong\u003e, 337\u0026ndash;350 (2007).\u003c/li\u003e\n\u003cli\u003eAlekseyenko, A. A., Gorchakov, A. A., Kharchenko, P. V. \u0026amp; Kuroda, M. I. Reciprocal interactions of human C10orf12 and C17orf96 with PRC2 revealed by BioTAP-XL cross-linking and affinity purification\u003cem\u003e. Proc. Natl. Acad. Sci. USA\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 2488\u0026ndash;2493 (2014).\u003c/li\u003e\n\u003cli\u003eAnggono, V., Smillie, K. J., Graham, M. E., Valova, V. A., Cousin, M. A. \u0026amp; Robinson, P. J. Syndapin I is the phosphorylation-regulated dynamin I partner in synaptic vesicle endocytosis. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 752\u0026ndash;760 (2006).\u003c/li\u003e\n\u003cli\u003eBanker, G. A. \u0026amp; Cowan, W. M. Rat hippocampal neurons in dispersed cell culture. \u003cem\u003eBrain Res.\u003c/em\u003e \u003cstrong\u003e126\u003c/strong\u003e (1977). \u003c/li\u003e\n\u003cli\u003eBartke, T., Vermeulen, M., Xhemalce, B., Robson, S. C., Mann, M. \u0026amp; Kouzarides, T. Nucleosome-interacting proteins regulated by DNA and histone methylation. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 470\u0026ndash;484 (2010).\u003c/li\u003e\n\u003cli\u003eBeringer, M., Pisano, P., Di Carlo, V., Blanco, E., Chammas, P., Viz\u0026aacute;n, P., Guti\u0026eacute;rrez, A., Aranda, S., Payer, B., Wierer, M. \u0026amp; Di Croce, L. EPOP Functionally Links Elongin and Polycomb in Pluripotent Stem Cells. \u003cem\u003eMol. Cell\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 645\u0026ndash;658 (2016).\u003c/li\u003e\n\u003cli\u003eBraun, A., Pinyol, R., Dahlhaus, R., Koch, D., Fonarev, P., Grant, B. D., Kessels, M. M. \u0026amp; Qualmann, B. EHD proteins associate with syndapin I and II and such interactions play a crucial role in endosomal recycling. \u003cem\u003eMol. Biol. Cell\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 3642-3658 (2005).\u003c/li\u003e\n\u003cli\u003eBrewer, G. J., Torricelli, J. R., Evege, E. K. \u0026amp; Price, P. J. Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination. \u003cem\u003eJ. Neurosci. Res\u003c/em\u003e. \u003cstrong\u003e35\u003c/strong\u003e, 567-576 (1993).\u003c/li\u003e\n\u003cli\u003eDe Cegli, R., Iacobacci, S., Flore, G., Gambardella, G., Mao, L., Cutillo, L., Lauria, M., Klose, J., Illingworth, E., Banfi, S. \u0026amp; di Bernardo, D. Reverse engineering a mouse embryonic stem cell-specific transcriptional network reveals a new modulator of neuronal differentiation. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 711\u0026ndash;726 (2013). \u003c/li\u003e\n\u003cli\u003eDharmalingam, E., Haeckel, A., Pinyol, R., Schwintzer, L., Koch, D., Kessels, M.M. \u0026amp; Qualmann, B. F-BAR proteins of the syndapin family shape the plasma membrane and are crucial for neuromorphogenesis. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 13315\u0026ndash;13327 (2009). \u003c/li\u003e\n\u003cli\u003eFischer, S., Weber, L. M. \u0026amp; Liefke, R. Evolutionary adaptation of the Polycomb repressive complex 2.\u003cem\u003e Epigenetics Chromatin\u003c/em\u003e. 15, 7 (2022). \u003c/li\u003e\n\u003cli\u003eGong, L., Liu, X., Yang, X., Yu, Z., Bhat, M. Y., Chen, S., Lemoff, A., Xing, C. \u0026amp; Liu, X. EPOP restricts PRC2.1 targeting to chromatin by directly modulating enzyme complex dimerization. \u003cem\u003eNat. Commun.\u003c/em\u003e 17, 1565 (2026).\u003c/li\u003e\n\u003cli\u003eGranat, J., Liu, S., Popoca, L., Oksuz, O. \u0026amp; Reinberg, D. EPOP and MTF2 activate PRC2 activity through DNA-sequence specificity\u003cem\u003e. Proc. Natl. Acad. Sci. USA \u003c/em\u003e\u003cstrong\u003e123\u003c/strong\u003e, e2527303123 (2026).\u003c/li\u003e\n\u003cli\u003eHaag, N., Schwintzer, L., Ahuja, R., Koch, N., Grimm, J., Heuer, H., Qualmann, B. \u0026amp; Kessels, M. M. The actin nucleator Cobl is crucial for Purkinje cell development and works in close conjunction with the F-actin binding protein Abp1. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 17842\u0026ndash;17856 (2012).\u003c/li\u003e\n\u003cli\u003eHealy, E., Mucha, M., Glancy, E., Fitzpatrick, D. J., Conway, E., Neikes, H. K., Monger, C., Van Mierlo, G., Baltissen, M. P., Koseki, Y., Vermeulen, M., Koseki, H. \u0026amp; Bracken, A. P.. PRC2.1 and PRC2.2 Synergize to Coordinate H3K27 Trimethylation. \u003cem\u003eMol. Cell\u003c/em\u003e \u003cstrong\u003e76\u003c/strong\u003e, 437\u0026ndash;452 (2019). \u003c/li\u003e\n\u003cli\u003eHenriquez, B., Bustos, F. J., Aguilar, R., Becerra, A., Simon, F., Montecino, M. \u0026amp; van Zundert, B. Ezh1 and Ezh2 differentially regulate PSD-95 gene transcription in developing hippocampal neurons. \u003cem\u003eMol. Cell. Neurosci.\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 130\u0026ndash;143 (2013).\u003c/li\u003e\n\u003cli\u003eHou, W., Izadi, M., Nemitz, S., Haag, N., Kessels, M.M. \u0026amp; Qualmann, B. The actin nucleator Cobl is controlled by calcium and calmodulin. \u003cem\u003ePLoS Biol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e1002233 (2015).\u003c/li\u003e\n\u003cli\u003eHou, W., Nemitz, S., Schopper, S., Nielsen, M.L., Kessels, M.M. \u0026amp; Qualmann, B. Arginine methylation by PRMT2 controls the functions of the actin nucleator Cobl. \u003cem\u003eDev. Cell\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 262\u0026ndash;275 (2018).\u003c/li\u003e\n\u003cli\u003eItoh, T., Erdmann, K. S., Roux, A., Habermann, B., Werner, H. \u0026amp; De Camilli, P. Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. \u003cem\u003eDev. Cell\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 791\u0026ndash;804 (2005).\u003c/li\u003e\n\u003cli\u003eIzadi, M., Schlobinski, D., Lahr, M., Schwintzer, L., Qualmann, B. \u0026amp; Kessels, M.M. Cobl-like promotes actin filament formation and dendritic branching using only a single WH2 domain. \u003cem\u003eJ. Cell Biol.\u003c/em\u003e \u003cstrong\u003e217\u003c/strong\u003e, 211-230 (2018).\u003c/li\u003e\n\u003cli\u003eIzadi, M., Seemann, E., Schlobinski, D., Schwintzer, L., Qualmann, B. \u0026amp; Kessels, M.M. Functional interdependence of the actin nucleator Cobl and Cobl-like in dendritic arbor development. \u003cem\u003eeLife\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e67718 (2021).\u003c/li\u003e\n\u003cli\u003eKessels, M.M. \u0026amp; Qualmann, B. Syndapins integrate N-WASP in receptor-mediated endocytosis. \u003cem\u003eEMBO J.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 6083\u0026ndash;6094 (2002).\u003c/li\u003e\n\u003cli\u003eKessels, M.M. \u0026amp; Qualmann, B. Syndapin oligomers interconnect the machineries for endocytic vesicle formation and actin polymerization. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e281\u003c/strong\u003e, 13285-13299 (2006).\u003c/li\u003e\n\u003cli\u003eKoch, N., Koch, D., Krueger, S., Tr\u0026ouml;ger, J., Sabanov, V., Ahmed, T., McMillan, L.E., Wolf, D., Montag, D., Kessels, M.M., Balschun, D. \u0026amp; Qualmann, B. Syndapin I loss-of-function in mice leads to schizophrenia-like symptoms. \u003cem\u003eCereb. Cortex\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 4306\u0026ndash;4324 (2020). \u003c/li\u003e\n\u003cli\u003eKoch, D., Spiwoks-Becker, I., Sabanov, V., Sinning, A., Dugladze, T., Stellmacher, A., Ahuja, R., Grimm, J., Sch\u0026uuml;ler, S., M\u0026uuml;ller, A., Angenstein, F., Ahmed, T., Diesler, A., Moser, M., tom Dieck, S., Spessert, R., Boeckers, T.M., F\u0026auml;ssler, R., H\u0026uuml;bner, C.A., Balschun, D., Gloveli, T., Kessels, M.M. \u0026amp; Qualmann, B. Proper synaptic vesicle formation and neuronal network activity critically rely on syndapin I. \u003cem\u003eEMBO J.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 4955-4969 (2011). \u003c/li\u003e\n\u003cli\u003eLiefke, R. \u0026amp; Shi, Y. The PRC2-associated factor C17orf96 is a novel CpG island regulator in mouse ES cells. \u003cem\u003eCell Discov.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 15008 (2015). \u003c/li\u003e\n\u003cli\u003eLiefke, R., Karwacki-Neisius, V. \u0026amp; Shi, Y. EPOP Interacts with Elongin BC and USP7 to Modulate the Chromatin Landscape. \u003cem\u003eMol. Cell\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 659\u0026ndash;672 (2016).\u003c/li\u003e\n\u003cli\u003eLiu, P. P., Xu, Y. J., Teng, Z. Q. \u0026amp; Liu, C. M. Polycomb Repressive Complex 2: Emerging Roles in the Central Nervous System.\u003cem\u003e Neuroscientist\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 208\u0026ndash;220 (2018).\u003c/li\u003e\n\u003cli\u003eMocavini, I., Mallol, A., Gutierrez, A., Chammas, P., Carretero, A., Dias, A., Blanco, E., Rodriguez Arce, I., Mallo, M., Di Croce, L. \u0026amp; Payer, B. The PRC2-associated factor EPOP is required for Hox gene regulation during axial development in mice. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e527\u003c/strong\u003e, 250\u0026ndash;259 (2025).\u003c/li\u003e\n\u003cli\u003eNolze, A., Schneider, J., Keil, R., Lederer, M., H\u0026uuml;ttelmaier, S., Kessels, M. M., Qualmann, B. \u0026amp; Hatzfeld, M. FMRP regulates actin filament organization via the armadillo protein p0071. \u003cem\u003eRNA\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 1483\u0026ndash;1496 (2013).\u003c/li\u003e\n\u003cli\u003eOkumura, F., Matsuzaki, M., Nakatsukasa, K. \u0026amp; Kamura, T. The Role of Elongin BC-Containing Ubiquitin Ligases. \u003cem\u003eFront. Oncol. \u003c/em\u003e\u003cstrong\u003e2\u003c/strong\u003e, 10 (2012). \u003c/li\u003e\n\u003cli\u003eQualmann, B., Roos, J., DiGregorio, P.J. \u0026amp; Kelly, R.B. Syndapin I, a synaptic dynamin-binding protein that associates with the neural Wiskott-Aldrich syndrome protein. \u003cem\u003eMol. Biol. Cell\u003c/em\u003e \u003cstrong\u003e10,\u003c/strong\u003e 501\u0026ndash;513 (1999).\u003c/li\u003e\n\u003cli\u003eQualmann, B., Boeckers, T. M., Jeromin, M., Gundelfinger, E. D. \u0026amp; Kessels, M. M. Linkage of the actin cytoskeleton to the postsynaptic density via direct interactions of Abp1 with the ProSAP/Shank family\u003cem\u003e. J. Neurosci.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 2481-2495 (2004).\u003c/li\u003e\n\u003cli\u003eQualmann, B. \u0026amp; Kelly, R.B. Syndapin isoforms participate in receptor-mediated endocytosis and actin organization. \u003cem\u003eJ. Cell Biol.\u003c/em\u003e \u003cstrong\u003e148,\u003c/strong\u003e 1047\u0026ndash;1062 (2000).\u003c/li\u003e\n\u003cli\u003eSchneider, K., Seemann, E., Liebmann, L., Ahuja, R., Koch, D., Westermann, M., H\u0026uuml;bner, C.A., Kessels, M.M., and Qualmann, B. (2014). ProSAP1 and membrane nanodomain-associated syndapin I promote postsynapse formation and function. \u003cem\u003eJ. Cell Biol.\u003c/em\u003e \u003cstrong\u003e205\u003c/strong\u003e, 197-215 (2014).\u003c/li\u003e\n\u003cli\u003eSchwintzer, L., Koch, N., Ahuja, R., Grimm, J., Kessels, M.M. \u0026amp; Qualmann, B. The functions of the actin nucleator Cobl in cellular morphogenesis critically depend on syndapin I\u003cem\u003e. EMBO J.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 3147\u0026ndash;3159 (2011).\u003c/li\u003e\n\u003cli\u003eSmits, A. H., Jansen, P. W., Poser, I., Hyman, A. A. \u0026amp; Vermeulen, M. Stoichiometry of chromatin-associated protein complexes revealed by label-free quantitative mass spectrometry-based proteomics. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, e28 (2013).\u003c/li\u003e\n\u003cli\u003eWolf, D., Hofbrucker-MacKenzie, S.A., Izadi, M., Seemann, E., Steiniger, F., Schwintzer, L., Koch, D., Kessels, M.M. \u0026amp; Qualmann, B. Ankyrin repeat-containing N-Ank proteins shape cellular membranes. \u003cem\u003eNat. Cell Biol.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 1191-1205 (2019). \u003c/li\u003e\n\u003cli\u003eZhang, Z., Jones, A., Sun, C. W., Li, C., Chang, C. W., Joo, H. Y., Dai, Q., Mysliwiec, M. R., Wu, L. C., Guo, Y., Yang, W., Liu, K., Pawlik, K. M., Erdjument-Bromage, H., Tempst, P., Lee, Y., Min, J., Townes, T. M. \u0026amp; Wang, H. PRC2 complexes with JARID2, MTF2, and esPRC2p48 in ES cells to modulate ES cell pluripotency and somatic cell reprogramming. \u003cem\u003eStem Cells \u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 229\u0026ndash;240 (2011).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9117199/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9117199/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRevealing functions of protein components distinct from initially described context significantly advances scientific knowledge. Here, we describe functions of EPOP (Elongin BC and Polycomb repressive complex 2-associated protein) distinct from EPOP\u0026rsquo;s established involvement in epigenetic regulation of gene expression in stem cells. These moonlighting functions that take place in a different cellular compartment, with a different binding partner and in mature, postmitotic cells. In neurons, we identified an EPOP subpool that was not nuclear but cytoplasmic and interacted with the membrane-shaping protein syndapin I. EPOP hereby showed a preference for the somatodendritic compartment and was also present in dendritic spines. \u003cem\u003eSyndapin I\u003c/em\u003e KO neurons revealed that EPOP\u0026rsquo;s extranuclear presence depended on its direct binding partner syndapin I. EPOP loss-of-function negatively impacted the density, morphology and postsynaptic-density composition of mushroom-type dendritic spines in mature hippocampal neurons and by closely phenocopying syndapin I loss-of-function phenotypes highlighted an intimate functional relationship of both components in postsynapses.\u003c/p\u003e","manuscriptTitle":"Identification of moonlighting, extranuclear and syndapin I-related functions of EPOP in mature neurons","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-30 06:44:29","doi":"10.21203/rs.3.rs-9117199/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-biology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commsbio","sideBox":"Learn more about [Communications Biology](http://www.nature.com/commsbio/)","snPcode":"","submissionUrl":"","title":"Communications Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1f0942ad-7a87-4237-a2d3-461ff4b36495","owner":[],"postedDate":"March 30th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65172948,"name":"Biological sciences/Neuroscience/Cellular neuroscience"},{"id":65172949,"name":"Biological sciences/Neuroscience/Development of the nervous system/Synaptic development"}],"tags":[],"updatedAt":"2026-04-28T16:56:05+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-30 06:44:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9117199","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9117199","identity":"rs-9117199","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}