Hippocampal Stimulation Promotes Intracellular Tip60 Dynamics with Concomitant Genome Reorganization and Synaptic Gene Activation.

Abstract

Genomic reorganizations mediating the engagement of target genes to transcription factories (TFs), characterized as specialized nuclear subcompartments enriched in hyperphosphorylated RNA polymerase II (RNAPII) and transcriptional regulators, act as an important layer of control in coordinating efficient gene transcription. However, their presence in hippocampal neurons and potential role in activity-dependent coregulation of genes within the brain remains unclear. Here, we investigate whether the well-characterized role for the histone acetyltransferase (HAT) Tip60 in mediating epigenetic control of inducible neuroplasticity genes involves TF associated chromatin reorganization in the hippocampus. We show that Tip60 shuttles into the nucleus following extracellular stimulation of rat hippocampal neurons with concomitant enhancement of Tip60 binding and activation of specific synaptic plasticity genes. Multicolor three-dimensional (3D) DNA fluorescent in situ hybridization (FISH) reveals that hippocampal stimulation mobilizes these same synaptic plasticity genes and Tip60 to RNAPII-rich TFs. Our data support a model by which external hippocampal stimulation promotes intracellular Tip60 HAT dynamics with concomitant TF associated genome reorganization to initiate Tip60 mediated synaptic gene activation.

1. Introduction

Neural networks of the brain can modify the efficacy and strength of their synaptic connections in response to external cues. This neurophysiological phenomenon is dependent upon the link between synaptic activity and gene expression and induces lasting changes in neuron excitability. At the molecular level, patterns of post-synaptic excitation trigger calcium-mediated activation of gene expression programs that give rise to synapse-specific morphological changes such as dendritic growth and branching and synaptogenesis (reviewed in (Greenberg et al. 1986; Bartel et al. 1989; Nedivi et al. 1993; Flavell and Greenberg 2008; Crepaldi et al. 2013). Within the hippocampus, this activity-dependent synaptic plasticity is the foundation for learning and memory consolidation (Fleischmann et al. 2003; Park et al. 2006; Madabhushi 2018).

The molecular mechanisms underlying the transcriptional response to depolarization involve the activation and recruitment of a variety of transcription factors and coregulators. Epigenetic gene control mechanisms, driven by histone-modifying enzymes, play an integral role in the regulation of activity-dependent gene expression (Peterson and Laniel 2004). Histone acetylation by histone acetyltransferases (HATs) like Tip60, leads to the decondensation of chromatin, generating transcriptionally favorable chromatin conformations for gene activation. While a compendium of evidence demonstrates the significance of histone acetylation induction in the modulation of gene transcription profiles following neuronal activation (Saunders et al. 2006; Barski et al. 2009; Saha et al. 2011; Xu et al. 2016), the specific HATs involved remain unclear. Our previous studies in Drosophila implicated Tip60’s HAT activity in the transcriptional regulation of neuroplasticity genes following the introduction of environmental stimuli (Xu et al. 2016). We further demonstrated that impairment of Tip60 HAT activity and/or improper Tip60 recruitment to this set of coregulated neuroplasticity genes triggers their epigenetic repression with concomitant disruption in learning and memory function (Panikker et. al. 2018). However, a fundamental question still to address is how does Tip60 acquire specificity towards multiple activity-dependent neuroplasticity gene loci? (Stilling et al. 2014). A clue to this question comes from studies demonstrating that in several mammalian cell types, RNA Polymerase II (RNAPII)-mediated transcription has been shown to occur at discrete intranuclear loci known as transcription factories (TFs) (Iborra et al. 1996; Osborne et al. 2004; Ragoczy et al. 2006; Osborne et al. 2007; Mitchell and Fraser 2008; Cook 2010; Chen 2017). In addition to the transcription machinery, TFs are enriched with a variety of transcriptional regulators and nuclear factors, making these regions transcriptional hubs optimized for efficient mRNA production (Schoenfelder et al. 2010; Melnik et al. 2011; Edelman and Fraser 2012). Previous findings have linked the simultaneous engagement of partner genes and shared TFs with their coordinated transcription in a variety of cell types (Reviewed in (Osborne et al. 2004; Osborne et al. 2007; Eskiw et al. 2010; Schoenfelder et al. 2010; Crepaldi et al. 2013). However, whether this process plays a role in the activation of activity-dependent transcriptional programs within the hippocampus remains relatively unexplored. Equally uncharacterized is the mélange of regulatory proteins orchestrating target gene association with TFs in response to synaptic activity.

Here we show that, within hippocampal neurons, the introduction of extracellular stimulation induces nuclear import of the HAT Tip60 with concomitant enhancement of Tip60 binding and activation of specific synaptic plasticity genes. Utilizing immuno-DNA fluorescence in situ hybridization (FISH), we visualized relocalization of these same activity-dependent gene loci to RNAPII-rich TFs in response to extracellular stimulation as well as enhanced co-association of Tip60 with the TFs harboring these activity-dependent genes. Our findings provide new fundamental insights into the cooperative role of epigenetic modifiers and transcription factories in the rapid and coordinated transcription of coregulated genes in response to neuronal activity within hippocampal neurons.

2. Materials and methods

2.1. Cell Culture

Rattus Norviegeus hippocampal neurons were prepared by the University of Pennsylvania, Mahoney Institute for Neuroscience Neuron Culture Service Center (http://www.med.upenn.edu/neuronsrus). Hippocampi of E18 male and female rats were dissected in ice-cold HBSS (Lonza), incubated in 0.25% Trypsin (Gibco) for 15 min at 37°C, washed with ice-cold HBSS twice for 5 min each, resuspended in DMEM (Bio Whittaker) supplemented with 10% FBS, and gently dissociated using a sterile Pasteur pipette. For immunofluorescence and immuno-DNA FISH experiments, neurons were plated on 12 mm round glass coverslips (Fisher), cultured in Neurobasal media (GIBCO) supplemented with B-27 (GIBCO) and incubated for 12 days. Neurons utilized for reverse transcriptase-real time polymerase chain reaction (RT-qPCR) and chromatin immunoprecipitation (ChIP) experiments were suspended (5×10⁶ to 10×10⁶ cells) in Neurobasal media supplemented with B-27 and utilized on the day of dissection. Neurons were stimulated with either 30 mM potassium chloride (KCl) or 50 μM N-methyl-D-aspartate (NMDA) dissolved in culture media for 30 min.

2.2. Immunofluorescence (IF)

Day in vitro (DIV) 12 hippocampal neurons cultured on glass coverslips were washed twice with PBS, fixed with 4% paraformaldehyde (Sigma) for 15 min at room temperature and permeabilized in 0.3% TritonX-100 (Fisher) in PBS for 5 min at room temperature. Cells were washed thrice with PBS for 5 min each followed by 5 min washes with TBS₅₀ (50 mM Tris, 150 mM NaCl, pH 7.4), and IF buffer (0.15 mM heat shock BSA, 0.15 mM protease-free BSA, TBS₅₀). After incubating in blocking buffer (2% BSA, 2% FBS in TBS₅₀) for 90 min, neurons were incubated overnight at 4°C with the following primary antibodies: Tip60 (Abcam, ab23886; 1:1000), Map2 (SCBT, sc-74421; 1:2000). Following 4 washes of 5 min each with IF buffer, neurons were incubated with anti-rabbit Alexa647 (Jackson Immuno, 111–607003; 1:300) or anti-mouse Cy3 (Abcam, ab97035; 1:200) secondary antibodies for 45 min at room temperature. Excess antibody was washed using successive 5 min washes with IF buffer, TBS₅₀, 2 mM MgCl₂ in × PBS, and 1× PBS. Coverslips were mounted and nuclei were stained using ProLong Gold Antifade with DAPI (Molecular Probes, P36941).

2.3. Imaging and quantification of Immunofluorescence data

Cells were viewed under confocal laser scanning microscopy (Olympus) and images were captured in a Z-stack series (0.2 μm) with FluoView acquisition software. Images were processed and analyzed using the National Institute of Health ImageJ software. Images acquisition and analysis were performed while blinded to the experimental condition. To quantify Tip60 localization pattern, the ratio of corrected total cellular fluorescence (CTCF) of nuclear to cytoplasmic Tip60 was measured. CTCF was calculated as integrated density-(area of the selected region of interest × mean fluorescence of background readings) (Gavet and Pines; Mccloy et al.).

2.4. RT-qPCR analysis

RNA from hippocampal neuron suspensions stimulated with KCl or left untreated were extracted using RNeasy Mini kit (Qiagen). 1 μg of total RNA was reverse-transcribed using the SuperScript II reverse transcriptase kit (Invitrogen), 10 mM dNTPs (New England Biolabs), and 0.2 mg/ml random hexamer primers (Roche Applied Science). Triplicate qPCR reactions were performed using the ABI 7500 Real-Time PCR system (Applied Biosystems) using a 20 μl reaction volume containing cDNA, 10 μl SYBR Green PCR Master Mix (Applied Biosystems), and 10 μM of each gene-specific forward and reverse primers. [Supplementary Table 1]. Thermocycler protocol consisted of an initial denaturation step at 95°C for 3 min, 45 cycles of amplification at 95°C for 15 sec and 60°C for 45 sec and followed by a dissociation stage 95°C for 15 sec, 60°C for 15 sec and 95°C for 15 sec. Fold change represents expression relative to unstimulated control cells and was calculated using the ΔΔCt method (Livak and Schmittgen 2001). Expression levels were normalized against reference gene Gapdh.

2.5. Chromatin Immunoprecipitation and qPCR analysis

Chromatin was extracted and sheared from hippocampal neuron suspensions stimulated with KCl or left untreated using truChIP™ Chromatin Shearing Kit (Covaris), following manufacturer’s instructions. Sonication was performed using a Covaris E220 Chromatin Shearer by applying 140 watts at 200 cycles/burst for 8 min resulting in chromatin fragments 200–700 bp in size, confirmed via agarose gel electrophoresis. Immunoprecipitation was performed using EZ-Magna ChIP A Chromatin Immunoprecipitation Kit (Millipore) with 20–30 μg sheared DNA incubated with 3 μg of Tip60 antibody (Abcam, ab23886) or without an antibody for Mock control samples, overnight at 4°C. Eluted and purified DNA from the immunoprecipitation was subjected to qPCR in 20 μl reactions containing 1 μl DNA, 0.25 μM of each gene-specific forward and reverse primers [Supplementary Table 1], and 10 μl SYBR Green PCR Master Mix. Triplicate qPCR reactions were performed. Thermocycler and its protocol were similar to RT-qPCR. Fold enrichment of the respective genes was calculated relative to the mock no antibody control (Panikker et. al. 2018).

2.6. Construction of probes for in situ hybridization

We performed immuno-DNA FISH with gene-specific, single-stranded DNA probes generated using asymmetrical PCR. DNA template for PCR was extracted from adult rat hippocampus tissue using Blood & Cell Culture DNA mini kit (Qiagen), following manufacturer’s instructions. Double-stranded copies of each target gene region were generated by standard PCR using DNA Taq Polymerase kit (Qiagen) with 1 ng - 1 μg genomic DNA, 200 μM standard dNTPs, and 0.5 μM forward and reverse primers [Supplementary Table 1]. Thermocycler protocol consisted of an initial denaturation step at 94°C for 3 min followed by 35 cycles of a three-step cycle consisting of a 1 min denaturation at 94°C, 30 sec annealing at 55°C (based on primer specification), and 1 min extension at 72°C and a final extension at 72°C for 10 min. Resulting double-stranded PCR products were purified using NucleoSpin Gel and PCR clean-up kit (Macherey-Nagel). Purified double-stranded DNA was then used as the template for the generation of single-stranded DNA FISH probes via asymmetrical PCR with the incorporation of modified nucleotides. 1 ng - 1 μg DNA was combined with 0.005 μM forward primer, 0.5 μM reverse primer, and 200 μM standard dNTPs mixed with aminoallyl-dUTP or biotin-dUTP. Thermocycler protocol consisted of an initial denaturation step at 94°C for 3 min followed by 35 cycles of a three-step cycle consisting of a 1 min denaturation at 94°C, 30 sec annealing at 55°C (based on primer specification), and 1 min extension at 72°C and a final extension at 72°C for 10 min. After amplification, the PCR product was electrophoresed through a 1% agarose gel, excised, and purified using phenol/chloroform extraction. Fluorophore conjugation was achieved by aminoallyl-modified and biotin-modified probes with NHS-Fluorescein (ThermoFisher) and streptavidin-Alexa568 (ThermoFisher), respectively. Excess fluorophore was eliminated using NucleoSpin Gel and PCR clean-up kit. Degree of fluorophore incorporation was measured using a spectrophotometer and calculated using Beer-Lambert equation with 5–8 dyes per 100 bp representing optimal incorporation.

2.7. Immuno-DNA Fluorescence in situ Hybridization (Immuno-DNA FISH)

Immuno-DNA FISH experiments were performed as previously described (MA AND TANESE 2013) with some modifications. DIV 12 hippocampal neurons were fixed for 20 min at room temperature with 4% paraformaldehyde. Cells were permeabilized with 0.25% TritonX-100 in 1× PBS for 5 min at room temperature followed by incubation with 100 μg/ml RNase A for 1 hr at 37°C. Prior to introduction, 20 ng FISH probes were pre-annealed with 1 μg salmon sperm DNA (ThermoFisher), 1 μg COT-1 DNA (Life Technologies), and 80% formamide in 1× SSC (Invitrogen) for 5 min at 65°C immediately before hybridization. Nuclear genomic DNA denaturation was obtained by incubating coverslips in 70% formamide in 2× SSC for 10 min at 80°C. 20 ng of the prepared probe was applied to each coverslip, incubated for 10 min at 80°C followed by overnight incubation in a humidity chamber at 37°C. Coverslips were extensively washed with PBS, blocked with blocking buffer (2% BSA, 2% FBS in TBS₅₀) and incubated with following primary antibodies: RNAPII S2P (Abcam, ab5095; 1:1000), Tip60 (SCBT, sc-74421; 1:50) for 2 hr at room temperature. Excess antibody was removed by washing with 0.02% TritonX-100 in 1× PBS. Detection was achieved by incubation with secondary antibodies for 1 hr at room temperature (anti-rabbit Alexa 647, Jackson ImmunoResearch 111–607003, 1:1000; anti-mouse Cy3, Abcam ab97035, 1:200). Following washes with 0.02% TritonX-100 in 1× PBS, coverslips were mounted, and nuclei were stained using ProLong Gold Antifade with DAPI (Molecular Probes).

2.8. Imaging and Quantification of Immuno-DNA FISH data

Cells were viewed under confocal laser scanning microscopy (Olympus) and images were captured in a Z-stack series (0.2 μm) with FluoView acquisition software. Images were processed and analyzed using the ImageJ software. Images acquisition and analysis were performed while blinded to the experimental condition.

To quantify locus position within the nucleus, a previously published approach for three-dimensional (3D) extrapolation from two-dimensional (2D) confocal microscopy images were used (Finch et al. 2008). A template consisting of six concentric circles derived from concentric spheres of linearly increasing volume (V = 1 to V = 6) was overlaid on each confocal image. The radius of each ring was calculated from the following equation: V = 4/3πr³. The location of each gene loci was quantified according to which shell (the area between two rings) the signal appeared and represented as the percentage of gene locus signals in each shell. Here, shell one defined as the outermost ring and shell 6 as the innermost ring.

To measure the colocalization of gene loci, TFs (RNAPII S2P), and Tip60, double labeling (respective gene & RNAPII S2P) and triple labeling (respective gene, RNAPII S2P, & Tip60) was quantified using the Image J software and represented as a percentage of colocalization in a blinded analysis. The percentages were calculated as (number of positive cells/total number of cells counted)*100.

2.9. Statistical analysis

For intracellular redistribution of Tip60, gene expression, and fold enrichment, two-tailed unpaired Student’s t-test with confidence intervals of 95% was used to calculate mean differences between groups. Data were tested for outliers using the Grubbs test. Statistical analysis was performed using the GraphPad Prism 6.0 (San Diego, CA, USA). Results are depicted as mean with standard errors (mean±SEM). Statistical significance was set at p < 0.05.

Whereas, two-sided Fisher’s exact test was used to measure the association between shell localization of gene loci and stimulation condition and association between gene loci co-localization with TFs/Tip60 and stimulation condition. Statistical analysis was performed using the GraphPad Prism 6.0 (San Diego, CA, USA). Results are depicted as a percentage. Statistical significance was set at p < 0.05.

3. Results

3.1. Extracellular stimulation leads to subcellular redistribution of Tip60 within hippocampal neurons.

Within neurons, Tip60’s HAT activity has been implicated in the modification of both cytosolic and nuclear proteins such as cell-cycle checkpoint regulatory proteins (Sun et al. 2005; Tang et al. 2006) and chromatin organizing histone proteins (Lorbeck et al. 2011), respectively. These intracellular functions are consistent with our previous findings of Tip60 localization in both the cytoplasm and nucleus in fly and mammalian neuronal cells (Sarthi and Elefant 2011; Xu et al. 2014; Xu et al. 2016). For example, we reported that Tip60 displays a nuclear-cytoplasmic distribution in Drosophila mushroom body and neuromuscular junction (NMJ) and further this distribution pattern is recapitulated in mammalian immature (DIV 6) rat hippocampal neurons and post-mortem human hippocampal tissue (Sarthi and Elefant 2011; Xu et al. 2014; Xu et al. 2016; Panikker 2018). Epigenetic modifiers such as histone deacetylases (HDACs) also display cytoplasmic/nuclear distribution patterns. Importantly, shuttling in and out of the nucleus following synaptic activity, provide these proteins the ability to carry out their epigenetic functions in a signal-mediated manner (Chawla et al. 2003). However, despite observations that Tip60 resides in both the nucleus and cytoplasm in neuronal cell types and contains a nuclear localization signal (NLS) and nuclear export signal (NES) within Tip60 (Xu et al. 2016), whether Tip60 possess similar stimulus-mediated cytoplasmic-nuclear shuttling capabilities as HDACs in fully matured hippocampal neurons remains unclear. To explore the molecular dynamics leading to activity-dependent gene expression, we utilized primary rat hippocampal neuron cultures of DIV 12 exposed to 30 mM KCl for 30 min. This stimulation paradigm is used to mimic synaptic activation in vitro by inducing neuronal depolarization, calcium influx, and gene transcription (Redmond et al. 2002; Lin et al. 2008; Nott et al. 2008). This short-term KCl exposure provides the opportunity to characterize the intracellular mechanisms that regulate early transcriptional events. DIV 12 hippocampal neurons are fully mature cultured neurons characterized by neurite maturation and dendrite stabilization (Barnes and Polleux 2009; Baj et al. 2014) and are appropriate for probing the molecular underpinnings regulating activity-dependent transcriptional modulation in mature neural circuits. Immunoblot analysis of proteins isolated from unstimulated and KCl stimulated DIV 12 rat hippocampal neurons revealed an increase in phosphorylation of CREB and ERK1 mediated well-characterized long-term potentiation (LTP) associated intracellular neuronal stimulatory activation pathways (Hardingham 2003; Luscher 2012), confirming the occurrence of typical neuronal stimulation [Supplementary Fig. 1].

To investigate a potential dynamic distribution of Tip60 within the cytosol and nucleus of hippocampal neurons, we utilized immunocytochemistry staining using antibodies against Tip60, the cytoplasmic marker Map2, and the nuclear marker DAPI. In the absence of stimulation, we observed Tip60 diffusely localized throughout the extranuclear and nuclear compartments, in agreement with our previous studies in other cell types (Xu et al. 2016) [Fig. 1A]. Higher-resolution analysis of the unstimulated neurons revealed greater levels of Tip60 in the soma, directly outside of the nucleus [Fig. 1B]. In response to stimulation, we observed a significant increase in the nuclear accumulation of Tip60 [Fig. 1C; t(133)=3.72, p < 0.001] suggesting a possible Tip60 nucleocytoplasmic shuttling in a stimulus-dependent manner, consistent with our previous studies in other cell types (Xu et al. 2016). Stimulation of these cells with NMDA, a well-characterized LTP associated glutamate receptor agonist involved in learning and memory, produced similar Tip60 nuclear import shuttling effects [Supplementary Fig. 2C; t(134)=7.34, p < 0.0001]. Notably, our findings make Tip60 the first HAT to demonstrate nuclear-shuttling capabilities in response to external cues. Taken together, these findings reveal the influence of extracellular stimulation via two external stimulatory cues linked to cognitive ability, on the nucleocytoplasmic shuttling of Tip60 within hippocampal neurons.

Fig. 1. Intracellular redistribution of Tip60 within hippocampal neurons following stimulation.

(A) Representative confocal images of immunocytochemical analysis of Tip60 localization in resting control DIV 12 rat hippocampal neurons. Cells were stained for Tip60 (red), cytoplasmic marker Map2 (green) and nuclei were counterstained with DAPI (blue). Immunostaining reveals cytoplasmic and nuclear distribution pattern for Tip60. (B) Representative confocal images of immunocytochemical analysis of the distribution of Tip60 within the cytoplasm and nucleus in unstimulated DIV 12 rat hippocampal neurons (top row) and after stimulation with 30 mM KCl for 30 min (bottom row). Cells were stained for Tip60 (red) and nuclei were counterstained with DAPI (blue). (C) Quantitative analysis of the ratio of corrected total cellular fluorescence (CTCF) of nuclear to cytoplasmic Tip60. Statistical significance was calculated using the two-tailed unpaired Student’s t-test. Data represents mean ± SEM, n = 68 per group and ***p < 0.001. Scale bars in A = 10 μm and B = 1 μm.

3.2. Tip60 is recruited to inducible neuronal genes following extracellular stimulation

Our previous studies in Drosophila implicated Tip60’s HAT activity in the transcriptional regulation of cognition-linked neuronal genes following the introduction of environmental stimuli (Xu et al. 2016). Subsequently, we demonstrated that a number of these genes are bona fide Tip60 HAT neuroplasticity targets whose expression is repressed in the brain of human Alzheimer’s disease (AD) Drosophila models and restored by increasing Tip60 HAT levels (Panikker 2018). This study also revealed that these same genes are repressed in the hippocampus of AD patients with concomitant Tip60 exclusion from the nucleus. Based on these findings, we hypothesized that the nuclear import of Tip60 in hippocampal cells might mediate its epigenetic control over activity-dependent cognitive-associated gene expression profiles. To test this hypothesis, we initially investigated whether the previously identified Tip60 neuroplasticity gene targets (Panikker 2018) are activated in hippocampal cells upon external stimulation. We focused on three Tip60 HAT targets [Map1a, Kcna2, Dvl2; Supplementary Fig. 3] with well-characterized roles in synaptic plasticity that we showed are repressed in the brain of AD-associated APP and Aβ42 Drosophila models and human AD patients. The RT-qPCR assays revealed that, as expected, each of these well-characterized activity-dependent genes undergoes rapid and significant transcriptional activation in primary rat hippocampal neurons upon exposure to 30 mM KCl for 30 min [Fig. 2A: Map1a t(16)=1.76, p = 0.097; Kcna2 t(12)=2.49, p < 0.05; Dvl2 t(15)=5.77, p < 0.05 & cFos t(11)=2.23, p < 0.05]. Approximate 2-fold induction of the well-characterized activity-dependent immediate early gene (IEG) cFos following KCl incubation further confirms neuronal stimulation and agrees with stimulus-mediated induction levels in other cell types (Osborne et al. 2007; Crepaldi et al. 2013).

Fig. 2. Increased Tip60 binding to activity-dependent synaptic plasticity genes in response to hippocampal neuron stimulation.

(A) RT- qPCR of four synaptic plasticity genes after a 30 min of 30 mM KCl exposure to hippocampal neuron culture. Gene expression values were normalized to Gapdh and calculated by the ΔΔCt method. (B) The histogram represents Tip60 binding enrichment (ChIP-qPCR) to synaptic plasticity genes after a 30 min of 30 mM KCl exposure to hippocampal neuron culture. An increased trend in Tip60 binding enrichment to indicated genes was observed to KCl exposure. Fold enrichment was calculated relative to the mock no antibody control. Statistical significance was calculated using the two-tailed unpaired Student’s t-test. Data represents mean ± SEM, n = 4–6 per group, *p < 0.05 and **p < 0.01.

Because dynamic changes in the binding of epigenetic modifiers to target genes represent a common feature of epigenetics-mediated gene control, we next asked whether Tip60 is recruited to the identified activity-dependent synaptic plasticity genes in response to depolarization. Performing Tip60 ChIP using chromatin isolated from KCl treated and untreated hippocampal neurons revealed a significant increase inTip60 binding to Map1a and an increasing trend in the proportion of Tip60 bound to Kcna2, Dvl2, and cFos in response to stimulation [Fig. 2B: Map1a t(7)=4.29, p < 0.01]. Our findings suggest that the stimulus-mediated nuclear import of Tip60 and its binding to co-regulated genes may facilitate Tip60 mediated epigenetic regulation of activity-dependent gene expression following hippocampal neuron activation.

3.3. Synaptic plasticity genes associate with transcription factories in response to extracellular stimulation

Visualization of mRNA synthesis within the mammalian nucleus revealed the presence of discrete sites enriched with transcription machinery and a collage of transcriptional regulatory proteins (Jackson et al. 1993; Iborra et al. 1996; Mitchell and Fraser 2008; Cook 2010; Schoenfelder et al. 2010; Melnik et al. 2011; Edelman and Fraser 2012). Previous studies in erythroid cells and cortical neurons have demonstrated that inducible genes dynamically localize to preformed TFs, thereby mediating stimulus-dependent transcription (Osborne et al. 2007; Crepaldi et al. 2013).

To examine the relationship between the dynamics of activity-dependent transcription, genomic reorganization and TF association within hippocampal neurons, we first sought to determine whether synaptic plasticity genes relocate within the hippocampal nucleus in response to stimulation. To address this question, we performed DNA FISH using custom single-stranded, fluorescently-labeled nucleic acid probes, highly specific to IEG and synaptic plasticity genes to directly visualize their location within the hippocampal nuclei of stimulated and unstimulated neurons. Each probe was generated using asymmetrical PCR with modified nucleotide incorporation on rat genomic template DNA followed by fluorophore conjugation and purification [Fig. 3A; protocol details can be found in Materials and Methods]. Agarose gel electrophoresis of each probe before fluorophore conjugation showed the expected base pair size and verified successful probe generation [Supplementary Fig. 4]. To quantify locus position within the nucleus, a previously published 3D extrapolation of 2D confocal microscopy images was used (Finch et al. 2008). Interestingly, we found that the localization of each synaptic plasticity gene was not fixed within the nucleus and appeared to change to different concentric rings following stimulation [Fig. 3C; Supplementary Table 2]. We observed a clear trend in the relocalization of gene loci to different shells in response to KCl for 3 genes (Map1a, Dvl2, cFos) and a significant relocalization for Kcna2 for shells 1 and 6 [Shell 1: Control 60% (6/10) and KCl 0% (0/13), p = 0.002; Shell 6: Control 0% (0/10) and KCl 46.2% (6/13), p = 0.019]. This suggests that synaptic plasticity genes are not stationary within the nucleus but rather can change their intranuclear position in response to external cues. This observation while novel to hippocampal neurons, is consistent with previous observations of gene loci dynamics following the introduction of stimulation in other cell types (Billia et al. 1992; Crepaldi et al. 2013; Walczak et al. 2013).

Fig. 3. Relocalization assessment of synaptic plasticity genes within the nucleus in response to KCl–mediated neuronal acitivation.

(A) Schematic illustrating single-stranded probe generation via asymmetrical PCR using modified nucleotides followed by fluorophore conjugation. (B) To quantify locus position within the nucleus, a previously published 3D extrapolation of 2D confocal microscopy images was used (FINCH et al. 2008). Representative image of locus position analysis in hippocampal neuron nuclei utilizing 6 shell template overlaid confocal FISH images. Each gene loci FISH signal was scored based on which shell the signal was found. The nuclear counterstain is DAPI (blue) and gene locus FISH probe is green. (C) The histogram represents the percentage of gene locus signals detected in stimulated (30 min of 30 mM KCl exposure) and unstimulated neurons found in each nuclear shell. A Fisher’s exact test was conducted to test the statistically significant association between shell localization of gene loci and stimulation condition. Here, *p < 0.05. n = 6–14 signals per group. See Supplementary Table 2 for details on analysis.

To examine this stimulus-mediated chromatin movement in relation to TFs, we utilized immuno-DNA FISH, a technique designed to allow the simultaneous detection of specific DNA loci and target proteins without compromising the 3D cytoarchitecture of the nucleus. This allowed for the direct visualization of synaptic plasticity gene localization in relation to TFs within stimulated and unstimulated hippocampal neurons. We directly visualized the location of each synaptic plasticity gene using the same highly specific fluorescently-labeled probes. Additionally, simultaneous immunostaining with an antibody recognizing the phosphorylated Ser2 residue of the C-terminal domain (CTD) of RNA Polymerase II (RNAPII S2P), the form of RNAPII associated with transcriptional elongation (Phatnani and Greenleaf 2006) that serves as a well-characterized and established marker for TFs (Alireza Ghamari 2013; Crepaldi et al. 2013), enabled detection of TFs within hippocampal neurons. We found that colocalization of each synaptic plasticity gene with TFs significantly increased in response to KCl-mediated extracellular stimulation, compared to unstimulated neurons [Fig. 4A, 4B: Map1a (p = 0.027); Kcna2 (p = 0.023); Dvl2 (p = 0.047) & cFos (p = 0.03) and Supplementary Table 3]. As a control, we also probed for the localization of Gapdh, a constitutively expressed neuronal gene, and found its TF localization frequency remained unchanged [Fig. 4B: Gapdh (p = 0.795)] between control [59.3% (16/27)] and KCl treated conditions [54.8% (17/31)]. Consistent with previous studies (Osborne et al. 2007; Crepaldi et al. 2013), these findings indicate an increased association of synaptic plasticity genes to transcription factories following stimulation in hippocampal neurons and may represent an integral step in activity-dependent regulation of gene transcription.

Fig. 4. Synaptic plasticity genes colocalize with TFs following hippocampal neuron stimulation.

(A) Representative confocal images of gene localization in rat DIV 12 hippocampal neurons in relation to TFs as visualized by triple-label immuno-DNA FISH. TFs detected by RNAPII S2P immunostaining (red), gene loci detected by custom single-stranded FISH probes (green) and nuclei counterstained with DAPI (blue). Enlarged images are selected gene loci (green, outlined in white). Scale bar indicates 1 μm. (B) The histogram represents the percentage of colocalization of synaptic plasticity gene and IEG (cFos) with TFs both in basal conditions and in response to KCl stimulation. Of note, and similar to prior publications (CREPALDI et al, 2013), in many representative images, one allele is observed as the other allele is often located within a different Z stack. Thus, multiple FISH signals per gene loci per condition are assessed. Data are represented as the percentage of colocalization [(number of positive cells/total number of cells counted)*100] and *p<0.05. A Fisher’s Exact test was conducted to test the statistically significant association between the colocalization of gene loci and TFs and stimulation condition. Here, *p < 0.05. n = 12 to 27 FISH signals per gene loci per condition. See Supplementary Table 3 for details on analysis.

3.4. Synaptic plasticity genes and Tip60 co-associate with the similar transcription factory

Recent findings have given rise to the concept that coregulated genes along with their regulatory protein(s) are assembled via genomic relocation to specialized TFs where they are simultaneously and efficiently transcribed. Supporting evidence for this model was experimentally provided by Schenfelder et al. (2010) who observed transcription factor Klf1-regulated mouse globin genes preferentially co-associated with factories harboring Klf1 (Schoenfelder et al. 2010). Further support of this theory was provided by Melnik et al., (2011) who subjected Hela cell factory isolates to mass spectrometry, revealing the presence of transcriptional coactivators, chromatin remodelers and histone methyltransferases (Melnik et al. 2011). Thus, we hypothesized that specialized TFs also exist in the brain that serves to coordinate expression of activity-dependent epigenetically controlled genes. To test this hypothesis, we examined the nuclear distribution of Tip60 in relation to the TFs harboring each synaptic plasticity gene locus by quadruple-label immuno-DNA FISH. We found that Tip60 was highly associated with each of the three synaptic plasticity genes we tested as well as with the IEG cFos within a given TF prior to stimulation [Fig. 5A & Fig. 5B; 75%–85% colocalization]. Notably, this is the first evidence of the colocalization of HATs with TFs in any cell type and is further evidence of the clustering of relevant regulatory proteins within these production-efficient hubs. To visually confirm our ChIP data [Fig. 2B] that had demonstrated a trend towards enhancement of Tip60 binding at synaptic gene loci following stimulation, we also assessed the localization of Tip60 with each gene following KCl stimulation using the quadruple-label immuno-DNA FISH [Fig. 5B]. Consistent with our ChIP findings, these immunohistochemical visualization studies reveal an increase of Tip60 association with Map1a, Dvl2, and cFos to 100% co-localization following the introduction of extracellular stimulation [Supplementary Table 4]. Collectively, these data support a model by which external hippocampal stimulation promotes intracellular Tip60 HAT dynamics with concomitant genome reorganization to promote epigenetic mediated synaptic gene activation.

Fig. 5. Synaptic plasticity genes colocalize with TFs harboring Tip60.

(A) Representative images of quadruple-label immuno-DNA FISH of DIV 12 rat hippocampal neurons under unstimulated and KCl stimulated conditions. Immunofluorescence detection of TFs by RNAPII S2P staining (red), Tip60 foci (yellow), gene loci detected by custom single-stranded FISH probes (green) and nuclei counterstained with DAPI (blue). Enlarged images show selected gene loci (green, outlined in white). Scale bar indicates 1 μm. (B) The histogram represents the percentage of genes within TFs found associated with Tip60 in hippocampal neurons under unstimulated and stimulated conditions as determined by quadruple-label immuno-DNA FISH. Of note, and similar to prior publications (CREPALDI et al, 2013), in many representative images, one allele is observed as the other allele is often located within a different Z stack. Thus, multiple FISH signals per gene loci per condition are assessed. Data are represented as the percentage of colocalization [(number of positive cells/total number of cells counted)*100]. A Fisher’s Exact test was conducted to measure significance in the association between colocalization of gene loci, TFs and Tip60, and stimulation condition. Here, n = 8 to 13 FISH signals per gene loci per condition. See Supplementary Table 4 for details on analysis.

4. Discussion

We previously demonstrated that Tip60, the second most highly expressed of the 18 different HATs in the mammalian brain (Stilling 2014), mediates acetylation-dependent synaptic plasticity gene activation in response to external cues (Xu et al. 2016), yet the mechanism underlying this process remained to be explored. Here we show a rapid nuclear import and binding of Tip60 to synaptic plasticity genes with concomitant transcriptional upregulation in response to KCl mediated hippocampal neuronal synaptic activation. Furthermore, with KCl stimulation, we observe a trend in the relocalization of synaptic plasticity genes within the nucleus and a significantly increased association of synaptic plasticity genes with RNAPII S2P marked TFs. Significantly, Tip60 is the first and only HAT thus far shown to undergo neuronal nuclear shuttling in response to external cues. Our observations introduce an exciting new model for how the extracellular environment influences HAT activity through the intracellular targeting of Tip60. Based on these findings, we propose a model in which stimulus-mediated nuclear import of Tip60 facilitates its association with target synaptic plasticity genes via gene re-localization and TFs association, thereby epigenetically-inducing gene expression profiles in response to hippocampal neuronal activation [Fig. 6]. Intriguingly, a compendium of recent studies reveals diminished nucleocytoplasmic transport in an array of neurodegenerative disorders that include amyotrophic lateral sclerosis, Huntington’s disease, and more recently Alzheimer’s disease, suggesting the nuclear shuttling deficits as a common theme underlying these debilitating cognitive disorders (Kim et. al. 2017). In support of this concept, we recently reported that Tip60 is predominantly excluded from the nucleus in human AD patients (Panikker et.al. 2018). Thus, we speculate that the observed stimulus-dependent nuclear import of Tip60 may be compromised in AD and attribute to cognitive deficits via epigenetic repression of Tip60 mediated neuroplasticity gene activation, a phenomenon we previously reported in the brain of two AD-associated Drosophila models (Xu et al. 2016; Panikker et.al. 2018).

Fig. 6. A model for Tip60’s role in the intracellular events regulating activity-dependent transcription within hippocampal neurons in response to extracellular stimulation.

(A) Under basal conditions, Tip60 is found both in the cytoplasm and the nucleus. Intranuclear Tip60 may be associated with certain synaptic plasticity genes prior to stimulation, representing a feature of rapidly inducible genes. (B) In response to KCl-mediated synaptic stimulation, Tip60 shuttles into the nucleus and is recruited to synaptic plasticity genes. Its HAT activity facilitates to locally unwind target gene regions, potentially exposing regulatory sequences and recruiting additional transcription factors and/or chromatin remodelers, mediating higher-order chromatin conformational changes. (C) The resulting chromatin reorganizations facilitate the recruitment of synaptic plasticity genes to RNAPII-rich TFs, thereby inducing their rapid, activity-dependent co-transcription.

The presence of transcription factors and coregulators bound to activity-dependent genes in the absence of stimulation is believed to enable the rapid induction of transcription upon a neuronal activity (Chen 2017) (reviewed in (Saunders et al. 2006; Barski et al. 2009). This concept is supported by the observation that TFIIIC, a transcription factor with acetyltransferase activity, associated with an upstream regulatory SINE sequence on stimulus-dependent genes before KCl-mediated cortical neuron stimulation (Crepaldi et al. 2013). Similarly, findings by Saha et al (2011) showed RNAPII docked on numerous IEG transcription start sites in the absence of stimulation, followed by rapid gene expression upon induction of cortical neuron activity, in vitro (Saha et al. 2011). These regulatory regions display patterns of transcriptionally favorable epigenetic modifications such as H3K9ac and H3K9K14ac before exposure to stimulation (Saha et al. 2011; Crepaldi et al. 2013). Consistent with these studies, here our Tip60-ChIP results reveal Tip60 enrichment at the IEG cFos within hippocampal neurons under unstimulated conditions that increase after KCl mediated external stimulation. Further, our immuno-DNA FISH immunohistochemistry experiments [Fig. 5] reveal co-localization of Tip60 with cFos (87.5%) and additional Tip60 neuroplasticity gene targets (71–77%) before stimulation and that this co-localization increases to 100% (except Dvl2) after KCl stimulation. These observations suggest that co-localization of Tip60 with its target genes prior to stimulation may play a similar role in maintaining some neuroplasticity genes in a transcription-ready state by being positioned before stimulation to induce acetylation and transcription of these first-order genes upon receipt of the appropriate signal. In line with this concept, several studies have reported similar findings of transcription factor and RNAPII binding to IEGs in the absence of stimulation and, therefore, may represent a feature of rapidly inducible genes (Saha et al. 2011; Crepaldi et al. 2013). Our work, together with previous findings (Xu et al. 2016; Panikker 2018; Chen 2019) suggest that Tip60 HAT action mediates the coordinated and timely expression of certain activity-dependent neuroplasticity genes within their associated TFs and may represent a feature of rapidly inducible genes.

A key feature of chromatin organization and transcriptional regulation involves the movement of genes within the nucleus (Reviewed in (Lanctot et al. 2007; Watson 2017). Utilizing DNA FISH, we show that neuronal activity mediates changes in the intranuclear localization of activity-dependent genes following neuronal activity in vitro. It has been widely argued that genes preferentially relocate from the periphery to interior of the nucleus upon transcriptional activation, ostensibly migrating from the heterochromatin-rich margins to the transcriptionally active interior. However, our data, except for Kcna2, is not consistent with this general assumption. We posit that the expression status of every gene might not entirely dependent on it locus positioning within the nucleus. This is in line with other studies reporting active gene expression occurring before moving away from the nuclear periphery or the localization of genes at the periphery irrespective of transcriptional state (Hewitt et al. 2004; Ragoczy et al. 2006; Williams et al. 2006). We suggest that of greater significance is the translocation of genes to transcriptionally active sites independent of their position within the nucleus, which occurs upon stimulation. We show that changes in the nuclear localization of inducible genes leads to their enhanced association with active RNAPII-enriched TFs and propose that this response directly results in their increased transcription following stimulation. Studies performed in other cell types have demonstrated that partner genes preferentially relocate to the same TF, thereby ensuring accurate and simultaneous transcription (Dekker et al. 2002; Osborne et al. 2004; Xu and Cook 2008; Schoenfelder et al. 2010; Crepaldi et al. 2013). Of note, each of the coregulated synaptic plasticity genes and IEG cFos examined in this study reside on a different chromosome of the rat genome [Supplementary Fig. 3]. Therefore, it is conceivable that neuronal stimulation would lead to clustering at shared TFs upon induction. Further examination into the co-localization of these partner genes with TFs would provide further insights into the underlying mechanism mediating their TF-mediated coordinated expression within hippocampal neurons. Our findings demonstrate that spatial repositioning of activity-dependent genes to TFs occurs within hippocampal nuclei and represents an important step in the coordinated transcription of coregulated, trans-positioned genes in response to external cues.

The interaction between regulatory gene regions such as promoter-enhancer interactions via transcription factor binding (Kieffer-Kwon et al. 2013; Zhang et al. 2013), results in the looping of the chromatin that is integral in mediating their genomic reorganization to TFs (Reviewed in (Cook 2010; Schoenfelder et al. 2010; Deng et al. 2013; Fitzpatrick et al. 2015). In line with this concept, previous mass-spectrometry analysis of TF isolates, revealed a heterogeneous amalgam of transcription factors and coactivators (Grande et al. 1997; Melnik et al. 2011). Here we show by utilizing ChIP in conjunction with quadruple-label immuno-DNA FISH, that Tip60 is co-localized within the same TFs as activity-dependent synaptic plasticity and IEG cFos within hippocampal neurons and that Tip60 is preferentially bound to at least some of these genes after stimulation. Our findings suggest that transcription factor-mediated genome spatial reorganization also involves the relocalization of synaptic plasticity genes to their respective TFs associated with Tip60 and potentially other epigenetic modifiers. Of note, direct manipulation of Tip60’s NLS and NES and Tip60 transcription factor binding partners will be required to tease apart the causative influence of Tip60 intracellular dynamics in synaptic plasticity gene activation.

5. Conclusions

Our findings here present a novel characterizational study of Tip60 intracellular and molecular dynamics within stimulated hippocampal neurons that may apply to other cell types. Our work supports a model by which KCl-mediated neuronal activation induces Tip60 nuclear import, a trend towards enhanced binding to activity-dependent genes, and activation of acetylation-driven chromatin decondensation leading to the increased association of genes and TFs thereby facilitating their transcription [Fig. 6]. This new understanding expands the portfolio of Tip60’s roles in regulating activity-dependent neuronal gene expression within the brain and points towards a new direction of research on the interplay between molecular epigenetics and specialized transcription factories.

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Ashley Karnay (amk456@drexel.edu).

Experimental Model and Subject Details

Source of cell lines used in this study is reported in the Key Resources Table.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Mouse monoclonal anti-Tip60	Santa Cruz Biotechnology	Cat#166323; RRID: AB_2296327
Rabbit polyclonal anti-Tip60	Abcam	Cat# ab23886; RRID: AB_778485
Mouse monoclonal anti-MAP-2	Santa Cruz Biotechnology	Cat#sc-74421; RRID: AB_1126215
Rabbit polyclonal anti-RNA polymerase II CTD repeat YSPTSPS (phospho S2)	Abcam	Cat# ab5095; RRID: AB_304749
Goat polyclonal anti-Rabbit IgG, Alexa Fluor 647	Jackson ImmunoResearch Labs	Cat# 111–607003; RRID: AB_2338084
Goat anti-Mouse IgG, Cy3	Abcam	Cat# ab97035; RRID: AB_10680176
Chemicals, Peptides, And Recombinant Proteins
Prolong Gold Antifade w/DAPI	Molecular Probes	P36941
NHS-Fluorescein (5/6-carboxyfluorescein succinimidyl ester), mixed isomer	Thermo Fisher	46410
Streptavidin-Alexa568	Thermo Fisher	S11226
AminoAllyl-dUTP	Thermo Fisher	FERR0091
Biotin-11-dUTP	Thermo Fisher	R0081
Critical Commercial Assays
EZ-Magna ChIP A Kit	EZ-Magna ChIP A Kit	EZ-Magna ChIP A Kit
TruChIP Chromatin Shearing Kit	TruChIP Chromatin Shearing Kit	TruChIP Chromatin Shearing Kit
Taq DNA Polymerase Kit	Taq DNA Polymerase Kit	Taq DNA Polymerase Kit
Blood & Cell Culture DNA Mini Kit	Blood & Cell Culture DNA Mini Kit	Blood & Cell Culture DNA Mini Kit
NucleoSpin Gel and PCR Clean-Up (50)	NucleoSpin Gel and PCR Clean-Up (50)	NucleoSpin Gel and PCR Clean-Up (50)
RNeasy Mini Kit	RNeasy Mini Kit	RNeasy Mini Kit
Superscript II Reverse Transcriptase Kit	Invitrogen	18064022
Experimental Models: Cell Lines
Rattus Norvegicus Hippocampal Cultures	University of Pennsylvania, Mahoney Institute for Nueroscience Neuron Culture Service Center	http://www.med.upenn.edu/neuronsrus
Oligonucleotides
Primers for RT-qPCR, see Table S1
Primers for ChIP, see Table S1
Primers for FISH Probe Generation, see Table S1
Software and Algorithms
ImageJ	NIH	https://imagej.net
Olympus fluoview acquisition software	Olympus, Center Valley, PA
IBM SPSS Statistics	IBM Analytics	https://www.ibm.com/analytics/data-science/predictive-analytics/spss-statistical-software
Corrected Total Cellular Fluorescence (CTCF)	McCloy, 2014; Gavet, 2010	http://doi.org.ezproxy2.library.drexel.edu/10.4161/cc.28401. http://doi.org.ezproxy2.library.drexel.edU/10.1016/j.devcel.2010.02.013

HIGHLIGHTS.

• Transcription factory (TF) function in activity-dependent neuroplasticity gene control in the brain is unclear

• Extracellular stimulation leads to subcellular redistribution of Tip60 within hippocampal neurons

• Tip60 is recruited to inducible neuroplasticity genes that are activated following extracellular stimulation

• Tip60 neuroplasticity gene targets associate with TFs in response to extracellular stimulation

• Neuroplasticity genes and Tip60 co-associate with the same TF following extracellular stimulation