Research ArticleNeuroscience

BDNF increases synaptic NMDA receptor abundance by enhancing the local translation of Pyk2 in cultured hippocampal neurons

See allHide authors and affiliations

Science Signaling  18 Jun 2019:
Vol. 12, Issue 586, eaav3577
DOI: 10.1126/scisignal.aav3577

An RNA binding protein in synaptic plasticity

Changes in neuronal activity, such as in response to the neurotrophin BDNF, occur partly through changes to the protein composition at the neuronal synapse. Afonso et al. found that the RNA binding protein hnRNP K was critical for the synaptic synthesis of the kinase PYK2 in hippocampal neurons upon BDNF stimulation. This synaptic localization of PYK2, in turn, promoted the formation and synaptic integration of GluN2B-containing NMDA receptors, which is important for the synaptic plasticity that underlies learning and memory formation. These findings not only pinpoint a critical kinase in the broader synaptic proteome for this phenomenon but also place an RNA binding protein at the core of its regulation.

Abstract

The effects of brain-derived neurotrophic factor (BDNF) in long-term synaptic potentiation (LTP) are thought to underlie learning and memory formation and are partly mediated by local protein synthesis. Here, we investigated the mechanisms that mediate BDNF-induced alterations in the synaptic proteome that are coupled to synaptic strengthening. BDNF induced the synaptic accumulation of GluN2B-containing NMDA receptors (NMDARs) and increased the amplitude of NMDAR-mediated miniature excitatory postsynaptic currents (mEPSCs) in cultured rat hippocampal neurons by a mechanism requiring activation of the protein tyrosine kinase Pyk2 and dependent on cellular protein synthesis. Single-particle tracking using quantum dot imaging revealed that the increase in the abundance of synaptic NMDAR currents correlated with their enhanced stability in the synaptic compartment. Furthermore, BDNF increased the local synthesis of Pyk2 at the synapse, and the observed increase in Pyk2 protein abundance along dendrites of cultured hippocampal neurons was mediated by a mechanism dependent on the ribonucleoprotein hnRNP K, which bound to Pyk2 mRNA and dissociated from it upon BDNF application. Knocking down hnRNP K reduced the BDNF-induced synaptic synthesis of Pyk2 protein, whereas its overexpression enhanced it. Together, these findings indicate that hnRNP K mediates the synaptic distribution of Pyk2 synthesis, and hence the synaptic incorporation of GluN2B-containing NMDARs, induced by BDNF, which may affect LTP and synaptic plasticity.

INTRODUCTION

The neurotrophin brain-derived neurotrophic factor (BDNF) is an important mediator of long-term synaptic potentiation (LTP) induced by high-frequency presynaptic stimulation in the hippocampus and in other brain regions (15). The early effects of BDNF are mediated by the posttranslational modification of synaptic components, whereas delayed responses require transcriptional activity and de novo protein synthesis (4, 5). Translation activity also mediates the facilitatory effects of BDNF signaling through its receptor tropomyosin receptor kinase B (TrkB) at CA1 synapses (48) and is reportedly required for the consolidation of LTP after infusion of BDNF in the dentate gyrus (9, 10).

The presence of the translation machinery at the synapse enables local protein synthesis from transcripts carried along dendrites in RNA granules (4, 11, 12), with a consequent rapid modification of the synaptic proteome in response to external stimuli. RNA granules travel along the microtubule tracks present in the dendritic compartment, and the stability of the transcripts during this transport is provided by the interaction with specific proteins (1214). BDNF-induced intracellular signaling was found to induce the disassembly of P-bodies in cultured hippocampal neurons, which may facilitate translation activity by making available the transcripts present in this class of RNA granules (15). Heterogeneous nuclear ribonucleoprotein K (hnRNP K), an RNA binding protein, interacts with a large number of transcripts coding for synaptic proteins and is an important mediator of the effects of BDNF on dendritic mRNA metabolism (16). Analysis of BDNF-induced alterations in the proteome of isolated synaptoneurosomes has enabled the identification of some of the diverse transcripts that are locally translated at the synapse in response to BDNF, such as Arc (8, 17) [a postsynaptic protein that plays an important role in LTP (18)], Homer2 [a postsynaptic density (PSD) scaffold protein], the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor subunit GluA1, and the Ca2+/calmodulin-dependent protein kinase II (CaMKII) (7, 8). BDNF also promotes the synaptic delivery of N-methyl-d-aspartate receptors (NMDARs) in cultured hippocampal neurons by a mechanism dependent on protein synthesis, but the molecular mechanisms involved remain to be elucidated (16).

NMDARs play an important role as mediators of synaptic plasticity mechanisms in the central nervous system (CNS), which underlie certain forms of learning and memory formation. Most NMDARs are tetrameric assemblies of two obligatory, glycine-binding GluN1 subunits and two glutamate-binding GluN2 subunits (19). GluN2A and GluN2B are the most abundant NMDAR subunits of this class in the forebrain and in the CA1 region of the hippocampus (2022), and influence the properties of the NMDAR of which they are apart. Compared with GluN2A-containing receptors, GluN2B-containing NMDARs exhibit greater Ca2+ permeability and charge transfer, lower opening probability and peak current, and slower deactivation, rise, and decay times (23). The ratio between GluN2A and GluN2B subunits is distinct between synapses and may change with development stage, in response to axonal input, and during axonal refinement that follows sensory experience (2426). These observations indicate that regulatory mechanisms facilitate changes in the synaptic content of GluN2-containing NMDAR to modulate their integrative capacity depending on the physiological context.

Here, we investigated the regulation of NMDAR by BDNF in rat hippocampal neurons. We found that BDNF stimulated the synaptic activity of NMDARs by increasing the proportion of receptors containing GluN2B subunits. This effect required the local synthesis of the nonreceptor tyrosine kinase Pyk2 and the participation of the RNA binding protein hnRNP K. Accordingly, stimulation of hippocampal neurons with BDNF promoted the dissociation of Pyk2 mRNA from hnRNP K. This mechanism may contribute to the protein synthesis–dependent effects of BDNF on LTP.

RESULTS

BDNF enhances the synaptic expression of GluN2B-containing NMDARs by a mechanism dependent on Pyk2 translation

BDNF was previously shown to increase the activity of postsynaptic NMDARs in cortical and hippocampal pyramidal neurons, in dentate gyrus granule cells (27, 28), and in cultured hippocampal neurons (16). We hypothesized that this effect could be mediated by the synaptic accumulation of GluN2B-containing NMDAR, because (i) this subunit is phosphorylated upon activation of BDNF-TrkB signaling (29, 30) and (ii) total surface expression of GluN2B-containing NMDARs increases after incubation with BDNF (31). The latter results were confirmed in immunocytochemistry experiments by live staining with an antibody against an extracellular epitope in the N-terminal region of GluN2B in cultured hippocampal neurons [days in vitro (DIV) 14 and 15] (Fig. 1A). After fixation and permeabilization of the plasma membrane, we stained for the neuronal marker microtubule-associated protein 2 (MAP2), and these preparations were analyzed for the number (Fig. 1B) and total intensity (Fig. 1C) of surface GluN2B puncta per dendritic length (based on MAP2 staining), as well as for the fluorescence intensity per puncta (Fig. 1D). Stimulation with BDNF (50 ng/ml) during 30 min increased the surface abundance of GluN2B-containing NMDARs, and this effect was abrogated when the cells were incubated with cycloheximide (50 μg/ml), which is commonly used to block protein synthesis.

Fig. 1 BDNF stimulation up-regulates the synaptic expression of GluN2B-containing NMDAR in a protein synthesis–dependent manner.

(A) Representative images of hippocampal neurons (DIV 14 and 15) that were preincubated with cycloheximide (CHX) (50 μg/ml) or vehicle (DMSO; 1:1000 dilution for 45 min) and then either maintained under the same conditions or stimulated with BDNF (50 ng/ml for 30 min), as indicated. Neurons were then live-immunostained for GluN2B using an antibody against an extracellular epitope in the GluN2B N terminus, fixed, and then further immunostained for PSD-95, vGlut1, and MAP2. Arrowheads indicate surface GluN2B–PSD-95–vGlut1–colocalized puncta. Scale bar, 5 μm. (B to G) Images described in (A) were analyzed for the total number (B) and intensity (C) of surface GluN2B puncta per dendritic length and GluN2B immunoreactivity per puncta (D). Synaptic (PSD-95– and vGlut1-colocalized) surface GluN2B number (E) and intensity (F) of puncta per density of excitatory synapses (number of puncta PSD-95–vGlut1 colocalized per dendrite length), as well as the immunoreactivity per synaptic puncta (G), were analyzed. Data are relative to the DMSO control and are the means ± SEM for the indicated number of neurons (n) in at least three independent experiments performed in different preparations. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way analysis of variance (ANOVA) with Bonferroni test.

To investigate whether BDNF treatment also increases the surface expression of GluN2B subunits specifically at synapses, and whether this effect would require protein synthesis, we compared the surface distribution of the NMDAR receptor subunit to that of the pre- and postsynaptic markers vesicular glutamate transporter 1 (vGluT1) and PSD-95, respectively, in addition to MAP2. The effect of BDNF on the surface abundance of synaptic GluN2B (Fig. 1A, arrowheads) was assessed by quantifying the number (Fig. 1E) and total intensity (Fig. 1F) of GluN2B signal puncta on the neuronal surface, as well as the immunoreactivity per puncta (Fig. 1G) that colocalized with PSD-95 and vGLUT1. BDNF increased the total number and total intensity of surface synaptic GluN2B puncta, and this effect was likewise blocked by cycloheximide (Fig. 1, E and F). Furthermore, BDNF up-regulated the number of synaptic surface GluN2B puncta by a mechanism that was partly insensitive to inhibition of protein synthesis (Fig. 1E), possibly involving activation of specific kinases and/or recruitment of receptors from extrasynaptic regions. Although BDNF enhanced the total intensity of synaptic surface GluN2B puncta along dendrites and increased the total number of puncta, the small increase in fluorescence intensity per synaptic puncta was not statistically significant (Fig. 1G). These results suggest that BDNF increases the number of synaptic puncta containing GluN2B NMDAR subunits.

The increased synaptic surface abundance of GluN2B upon stimulation of hippocampal neurons with BDNF suggests that incubation with the neurotrophin may increase NMDAR-mediated activity at the synapse. To address this hypothesis, we tested the effect of BDNF on the amplitude and frequency of NMDAR-mediated miniature excitatory postsynaptic currents (mEPSCs). Alterations in the number of NMDAR at the synapse are expected to correlate with changes in the amplitude of mEPSC, whereas changes in the number of synapses containing NMDAR and/or presynaptic effects of the neurotrophin are coupled to alterations in the frequency of mEPSCs. Previous studies suggested that the BDNF-induced up-regulation in the frequency of NMDAR-mediated mEPSCs in cultured hippocampal neurons is mediated by a postsynaptic mechanism (16). Incubation of hippocampal neurons with BDNF (50 ng/ml for 30 to 40 min) increased the amplitude of mEPSCs, and this effect was abrogated by the GluN2B inhibitor conantokin G (Fig. 2, A and B) (32). Furthermore, conantokin G significantly decreased the amplitude of NMDAR-mediated mEPSCs in hippocampal neurons under control conditions. These results show the contribution of GluN2B-containing receptors to NMDAR-mediated mEPSCs, in accordance with the evidence obtained in our immunocytochemistry experiments above. Inhibition of GluN2B with conantokin G also abrogated the BDNF-induced up-regulation in the frequency of NMDAR-mediated mEPSCs (Fig. 2, C and D). These results suggest that BDNF may, at least in part, enhance NMDAR-mediated synaptic transmission by recruiting GluN2B-containing receptors to synapses previously lacking this type of receptors. The conantokin G–insensitive mEPSCs, which were not altered upon stimulation with BDNF, may be mediated by NMDAR expressing GluN2A subunits.

Fig. 2 BDNF treatment increases the abundance of functional synaptic GluN2B-containing NMDARs in cultured hippocampal neurons.

(A to D) NMDAR mEPSCs were recorded in cultured rat hippocampal neurons after incubation in culture medium under control conditions or in the presence of BDNF (50 ng/ml) for 30 to 40 min. mEPSCs were recorded in a Na+ salt solution, and where indicated, the medium was supplemented with conantokin G (Con G; 3 μM). The average mEPSC traces recorded are shown in (A), and representative traces are shown in (C). Data (B and D) are mean ± SEM mEPSC amplitude and frequency, respectively, for the indicated number of neurons (n) from at least three independent preparations. **P < 0.01, ****P < 0.0001 by one-way ANOVA with Bonferroni test.

Previously reported data suggest that NMDAR synaptic content depends on the equilibrium between intracellular trafficking and rapid lateral diffusion of the receptor within the synaptic area (33). To determine whether the increased synaptic GluN2B expression upon stimulation of hippocampal neurons with BDNF depends on alterations of NMDAR surface dynamics, we directly investigated GluN2B-containing NMDAR surface diffusion using single-particle tracking (SPT) (34). This method allows measuring the diffusion properties of single labeled receptors over time. To track individual receptors, quantum dots (QDs) were coupled to antibodies directed against an extracellular epitope of the GluN2B subunit before incubation with cultured hippocampal neurons. Live imaging experiments, to analyze diffusion properties of GluN2B-containing NMDAR, showed an increased percentage of immobile receptors at the synapse after stimulation with BDNF for 30 min (Fig. 3, A and B). Moreover, under the same conditions, there was a decrease in the mean squared displacement (MSD) of synaptic and extrasynaptic GluN2B-containing NMDAR (Fig. 3, C and D), as well as in the respective diffusion coefficient (Fig. 3, E and F). Together, the observed alterations in the diffusion of GluN2B subunits suggest that BDNF increases the synaptic content of these NMDAR subunits by reducing their surface mobility.

Fig. 3 GluN2B-NMDAR surface dynamics is altered by BDNF.

(A) Representative trajectories of extrasynaptic (blue lines) and synaptic (red lines) GluN2B-NMDARs in hippocampal neurons transfected with PSD95-FingR-GFP (green fluorescent protein) at DIV 10 and 4 days later incubated with or without BDNF (50 ng/ml for 30 min). The cells were then incubated for 3 min in a solution containing QD particles coupled to a primary antibody against GluN2B-NMDAR subunits. Imaging captured 200 frames with a 50-ms acquisition frequency. Scale bar, 1 μm. (B) Percentage of extrasynaptic and synaptic immobile GluN2B-NMDAR trajectories in neurons described in (A). ns, not significant. (C and D) MSD over time of the extrasynaptic (C) and synaptic (D) surface GluN2B-NMDAR trajectories in hippocampal neurons described in (A). (E and F) Comparison of GluN2B-NMDAR instantaneous diffusion coefficient (μm2 s−1) within the extrasynaptic (E) and PSD area (F) in neurons described in (A). Data are means ± SEM from five independent experiments performed in distinct preparations. *P < 0.05, ****P < 0.0001 by unpaired Student’s t test.

De novo synthesis of Pyk2 mediates the effects of BDNF on the synaptic expression of GluN2B-containing NMDAR

Pyk2 and Src family kinases (SFKs) are part of the NMDAR complex (35, 36), and it was proposed that Ca2+-dependent activation of Pyk2 plays a role in the regulation of NMDAR and synaptic plasticity by a mechanism involving SFKs (37, 38). Therefore, we hypothesized that this pathway could play a role in the modulation of synaptic NMDAR by BDNF, downstream of local synthesis of Pyk2. The role of Pyk2 in BDNF-induced increase in the surface abundance of GluN2B-containing NMDAR was investigated in cultured hippocampal neurons using each of two short hairpin RNAs (shRNAs) to knock down Pyk2 (shA2-Pyk2 and shA4-Pyk2). We first validated the efficiency of these shRNAs in the C6 cell line and in neuronal primary cultures (fig. S1). Knockdown of Pyk2 abrogated the effects of BDNF on the total surface abundance of GluN2B subunit in the dendritic compartment, as determined by live immunostaining with an antibody against an extracellular epitope in the N-terminal region of the protein and by colocalization with the dendritic marker MAP2 (Fig. 4, A to D). We then investigated whether Pyk2 also plays a role in the BDNF-induced up-regulation of GluN2B surface expression at the synaptic compartment, as determined by colocalization with the presynaptic marker vGlut1. BDNF induced a significant increase in the number (Fig. 4, A and E) and intensity (Fig. 4, A and F) of synaptic surface GluN2B puncta in neurons transfected with the control (scrambled) shRNA, whereas knockdown of Pyk2 abolished these effects of the neurotrophin. In accordance with our above results (Fig. 1G), the fluorescence intensity per puncta was not significantly changed upon stimulation with BDNF (Fig. 4G). Together, these results show a role for Pyk2 in the up-regulation of postsynaptic GluN2B-NMDAR after acute treatment with BDNF. Furthermore, the neurotrophin enhanced the percentage of synapses containing surface GluN2B in cultured hippocampal neurons by a mechanism dependent on Pyk2 (Fig. 4, A and H). In accordance with the results obtained in our immunocytochemistry experiments, down-regulation of Pyk2 abrogated the BDNF-evoked increase in the amplitude of NMDAR-mediated mEPSCs but was without effect on the currents measured under control conditions (Fig. 5, A and B). Down-regulation of Pyk2 also reduced the effects of BDNF on the frequency of total NMDAR-mediated mEPSCs (Fig. 5, C and D). Together, these results indicate that Pyk2 mediates the BDNF-induced increase in surface synaptic GluN2B-containing NMDAR, and this accompanies the overall increase in the surface abundance of this receptor subunit evoked by BDNF.

Fig. 4 BDNF-induced increase in the synaptic surface expression of NMDAR-containing GluN2B subunits requires Pyk2.

(A) Representative images of rat hippocampal neurons transfected with sh1-Scramble (sh1-Scrbl) or one of two Pyk2-targeted (shA2-Pyk2 or shA4-Pyk2) shRNA at DIV 12. At DIV 15, cultures were then either maintained under the same conditions or stimulated with BDNF (50 ng/ml for 30 min) as indicated. Neurons were live-immunostained for GluN2B using an antibody against an extracellular epitope located in the GluN2B N terminus, fixed and permeabilized, and then further immunostained for vGLUT1, GFP, and MAP2. Scale bar, 5 μm. (B to H) Images represented in (A) were analyzed for the number (B) and intensity (C) of surface GluN2B puncta per dendritic length, as well as GluN2B immunoreactivity per puncta (D), quantified relative to those in the sh1-Scramble control condition. The number (E) and intensity (F) of synaptic (vGlut1-colocalized) surface GluN2B puncta per density of excitatory synapses (number of puncta PSD-95–vGlut1 colocalized per dendrite length), the immunoreactivity per synaptic puncta (G), and the percentage of synapses containing surface GluN2B (number of vGluT1 puncta colocalized with surface GluN2B/total vGluT1 number of puncta) (H) were also analyzed. Data are means ± SEM for the indicated number of neurons (n) from at least four independent experiments, performed in independent preparations. ***P < 0.001, ****P < 0.0001 between indicated conditions; ##P < 0.01, ####P < 0.0001 versus scramble control by one-way ANOVA with Bonferroni test.

Fig. 5 The up-regulation of NMDAR-mediated mEPSCs by BDNF is mediated by Pyk2.

(A to D) Rat hippocampal neurons were either untransfected (Ctrl) or transfected with sh1-Scramble (Sh1-Scrbl) or Sh4-Pyk2 at DIV 12 and, at DIV 15, were then either maintained under control conditions or stimulated with BDNF (50 ng/ml) for 30 to 40 min. Average NMDAR-mediated mEPSC traces recorded at −60 mV are shown in (A), and representative traces are shown in (C). Mean ± SEM amplitude (B) and frequency (D) of NMDAR-mediated mEPSCs were quantified. *P < 0.05, **P < 0.01, ***P < 0.001 by one-way ANOVA with Bonferroni test.

Pyk2 is a kinase that is widely expressed in the CNS, and its autophosphorylation on Tyr402 allows the interaction with the SH2 domain of Src. This interaction leads to the phosphorylation of Pyk2 in the kinase domain activation loop, which is required for maximal activity (39). To determine whether BDNF enhances Pyk2 activity at the synapse, Western blot experiments were performed using a phosphospecific antibody to analyze the phosphorylation of the kinase in adult rat (10 to 12 weeks) hippocampal synaptoneurosomes, a subcellular fraction containing the pre- and postsynaptic regions (17). BDNF stimulation for 20 to 30 min increased the abundance of phosphorylated Pyk2, as determined after normalization with both β-tubulin and total Pyk2 abundance (Fig. 6, A to C). To determine whether the effects of BDNF on Pyk2 phosphorylation were specifically observed at the synapse, we performed similar experiments in total extracts of cultured hippocampal neurons. In this case, a 20-min stimulation with BDNF did not change total Pyk2 protein levels, and no alterations in the phosphorylation of the kinase were observed (Fig. 6, D and E). Similarly, no significant alterations were observed in the pPyk2/total Pyk2 ratio (Fig. 6, D and F). Together, these results indicate that Pyk2 undergoes a rapid and transient activation at the synapse after stimulation with BDNF by a mechanism involving, at least in part, dendritic Pyk2 synthesis.

Fig. 6 BDNF-induced increase in synaptic expression of GluN2B-containing NMDAR is mediated by activation of Pyk2.

(A to C) Representative Western blots (A) and analysis (B and C) of synaptic, phosphorylated (Tyr402) Pyk2 abundance in hippocampal synaptoneurosomes that were first warmed for 5 min at 30°C and then either unperturbed or stimulated with BDNF (50 ng/ml) for 10, 20, or 30 min. (D to F) As in (A) to (C) in cultured hippocampal neurons (high-density cultures; DIV 14 and 15) either maintained under control conditions or stimulated with BDNF (50 ng/ml) for 20 min. Data are means ± SEM for the number of independent experiments indicated (n). *P < 0.05, **P < 0.01, ***P < 0.001 by unpaired Student’s t test. (G) Representative images of hippocampal neurons that were transfected with wild-type (WT) or phospho-mutant, kinase-deficient Pyk2 (Y402F) at DIV 12, and at DIV 15 were maintained under control conditions or stimulated with BDNF (50 ng/ml) for 30 min, live-immunostained for GluN2B (using an antibody against an extracellular epitope in the GluN2B N terminus), fixed and permeabilized, and further immunostained for Flag (transfection marker) and MAP2. Scale bar, 5 μm. (H to J) Analysis of total number (H) and intensity (I) of surface GluN2B puncta per dendritic length, as well as for GluN2B immunoreactivity per puncta (J), in neurons described in (G). Data were quantified relative to the empty vector control (Flag-Empty) and are means ± SEM for the indicated number of neurons (n) from at least three independent experiments. *P < 0.05, ***P < 0.001 between the indicated conditions; #P < 0.05, ##P < 0.01, ###P < 0.0001 versus empty control by one-way ANOVA with Bonferroni test.

To determine whether BDNF-induced increase in GluN2B surface expression is mediated by activation/phosphorylation of Pyk2, the effect of the neurotrophin was investigated in hippocampal neurons transfected with a phospho-mutant form of Pyk2 (Pyk2-Y402F), which has no kinase activity, or with the wild-type form of Pyk2 (Pyk2-WT). As a control, hippocampal neurons were transfected with the Flag vector (Flag-Empty). Transfection of hippocampal neurons with the phospho-mutant form of Pyk2 abrogated the BDNF-induced up-regulation in the dendritic surface expression of GluN2B, as determined by live immunostaining with a GluN2B antibody against an extracellular epitope located at the N terminus of the protein (Fig. 6, G to J). Overexpression of the phospho-mutant Pyk2 (Y402F) also decreased the surface expression of GluN2B under resting conditions (Fig. 6, G to J). These results indicate that activation/phosphorylation of Pyk2 is required both for the maintenance of basal GluN2B expression and for its increase upon BDNF stimulation. Unexpectedly, overexpressing Pyk2-WT mimicked the effects of the neurotrophin on surface abundance of GluN2B-NMDARs. These findings suggest that the role of BDNF in the regulation of surface expression of these receptors is mediated not only by activating Pyk2 but also by promoting the dendritic/synaptic accumulation of the kinase. Furthermore, BDNF treatment did not further increase surface GluN2B expression in neurons overexpressing Pyk2 (Fig. 6, G to J), further supporting this hypothesis.

Together, these results indicate that BDNF increases GluN2B total surface expression by two mechanisms: (i) up-regulation of the dendritic/synaptic levels of Pyk2, which increases the surface expression of GluN2B-containing NMDARs, and (ii) activation of signaling mechanisms coupled to activation of Pyk2 phosphorylation, which is required to enhance GluN2B surface expression. Pyk2 activity is also required to maintain GluN2B surface expression under resting conditions.

BDNF induces the local synthesis of Pyk2 at the synapse

To further investigate the effect of BDNF on local protein synthesis in the dendritic compartment, and, in particular, at the synapse, we used the FUNCAT-PLA (fluorescent noncanonical amino acid tagging–proximity ligation assay) method that allows monitoring specific newly synthesized proteins with spatial resolution (40). Synthesis of Pyk2 in dendrites was evaluated by colocalization with the dendritic marker MAP2 (Fig. 7A). Incubation of hippocampal neurons with BDNF (50 ng/ml) for 30 min significantly increased the number (Fig. 7B) and total intensity (Fig. 7C) of Pyk2 FUNCAT-PLA puncta, whereas the signal intensity per puncta was not significantly changed (Fig. 7D). Furthermore, incubation of hippocampal neurons with BDNF up-regulated the number of synaptic Pyk2 FUNCAT-PLA puncta, as determined by colocalization with PSD95 (Fig. 7E). An increase in the total intensity of synaptic Pyk2 FUNCAT-PLA was also observed under the same conditions, but the effect was not statistically significant (Fig. 7F). In contrast, no alteration was observed in the intensity of the synaptic Pyk2 FUNCAT-PLA signal per puncta, indicating that the effects of BDNF on Pyk2 translation are specific for a subpopulation of synapses not labeled under resting conditions.

Fig. 7 BDNF increases the synaptic synthesis of Pyk2.

(A) Representative images of hippocampal neurons (DIV 14 and 15) that were incubated with azidohomoalanine (AHA) (4 mM) in the presence or absence of BDNF (50 ng/ml) for 30 min, fixed, and then subjected to the click chemistry technique (FUNCAT) to biotinylate AHA. Antibodies against biotin and Pyk2 were used to detect close proximity between Pyk2 and newly synthetized proteins. The FUNCAT-PLA signal (green) was obtained using PLAminus and PLAplus oligonucleotides coupled to secondary antibodies, together with detection reagents for ligation and amplification. PLA signal was amplified through binding of fluorescent detection probes. Scale bar, 5 μm. Arrowheads indicate PLA–PSD-95–colocalized puncta. (B to G) PLA neurons described in (A) were immunostained for PSD-95 and MAP2 and analyzed for the total number (B) and intensity (C) of PLA puncta per dendritic length, as well as for the immunoreactivity per puncta (D). The number (F) and intensity (G) of synaptic (PSD-95–colocalized) PLA puncta per dendrite length and the immunoreactivity per synaptic puncta (G) were also analyzed. Data were quantified relative to control and are means ± SEM for the indicated number of neurons (n) in three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001 by unpaired Student’s t test.

The BDNF-induced increase in local translation of Pyk2 at the synapse was further investigated by measuring the effect of the neurotrophin on total Pyk2 protein abundance in hippocampal synaptoneurosomes. BDNF (20 min) increased synaptic Pyk2 protein abundance, and this effect was abrogated in the presence of cycloheximide (Fig. 8, A to C).

Fig. 8 BDNF treatment increases local protein translation and up-regulates Pyk2 protein levels in hippocampal synaptoneurosomes.

(A) Timeline of the protocol used to test the effect of BDNF on Pyk2 protein abundance in hippocampal synaptoneurosomes. (B and C) Representative Western blot (B) and analysis (C) of Pyk2 abundance in hippocampal synaptoneurosomes that underwent the protocol described in (A). Data are means ± SEM from four or five independent experiments as indicated (n). Densitometry of Pyk2 bands was normalized to that of β-tubulin. Statistical significance was calculated using the one-way ANOVA (P < 0.0001) with Bonferroni’s multiple comparison test: ***P < 0.001, ****P < 0.0001. (D and E) As described for (B) and (C) in cultured hippocampal neurons. Statistical analysis was performed using Student’s t test.

The results described above show an effect of BDNF in the up-regulation of Pyk2 in the synaptic compartment in hippocampal neurons. To determine whether BDNF also induces an overall increase in Pyk2 protein synthesis in cultured hippocampal neurons, the cells were stimulated with BDNF for 20 min and Pyk2 protein levels were evaluated by Western blot. In contrast with the results obtained in hippocampal synaptoneurosomes, BDNF did not affect Pyk2 total protein abundance in hippocampal neurons (Fig. 8, D and E), suggesting a specific compartmentalized effect of BDNF, which is not reflected in the total abundance of the protein. Together, these results show that BDNF induces a specific increase in synaptic Pyk2 mediated by de novo local protein synthesis.

BDNF up-regulates synaptic Pyk2 protein abundance by a mechanism dependent on hnRNP K

To further investigate the effects of BDNF on synaptic Pyk2 protein abundance, cultured hippocampal neurons (14 and 15 DIV) were stimulated with the neurotrophin (50 ng/ml) for 30 or 60 min, and the dendritic distribution of the kinase was evaluated using immunocytochemistry by colocalization with MAP2. The fluorescence images obtained were analyzed for the number (Fig. 9, A and B) and intensity (Fig. 9, A and C) of Pyk2 puncta per dendritic length (marked with MAP2), as well as for Pyk2 immunoreactivity per puncta (Fig. 9D). Stimulation with BDNF for 30 or 60 min enhanced to a similar extent the number and intensity of Pyk2 puncta, and this effect was abrogated in the presence of cycloheximide (50 μg/ml). Pyk2 labeling per puncta was also statistically significantly increased in neurons stimulated with BDNF for 30 min, but no significant effect was observed for longer incubation periods.

Fig. 9 BDNF up-regulates synaptic Pyk2 protein abundance and the percentage of synapses containing Pyk2 in a protein synthesis–dependent manner.

(A) Representative images of rat hippocampal neurons maintained under control conditions (Ctrl) or incubated with BDNF (50 ng/ml) for 30 or 60 min in the presence or absence of cycloheximide (50 μg/ml), as indicated. Neurons were live-immunostained for GluN2B using an antibody against an extracellular epitope located in the GluN2B N terminus, fixed and permeabilized, and then further immunostained for Pyk2, MAP2, and PSD-95. Hippocampal neurons were preincubated with cycloheximide (50 μg/ml) or vehicle (DMSO, 1:1000 dilution) for 45 min before stimulation with BDNF, as indicated. Arrowheads indicate Pyk2–PSD-95–colocalized puncta. Scale bars, 5 μm. (B to I) Images represented in (A) were analyzed for the number (B) and intensity (C) of Pyk2 puncta per dendritic length, as well as for Pyk2 immunoreactivity per puncta (D), quantified as percentage of the respective DMSO control (at 30 or 60 min). The number (E) and intensity (F) of synaptic (PSD-95–colocalized) Pyk2 as well as Pyk2 immunoreactivity per puncta (G) were also analyzed. The percentage of Pyk2 showing a synaptic distribution (number of Pyk2 puncta colocalized with PSD-95/total Pyk2 number of puncta) and the percentage of synapses containing Pyk2 (number of PSD-95 puncta colocalized with Pyk2/total PSD-95 number of puncta) are shown in (H) and (I), respectively. Data are average ± SEM for the indicated number of neurons (n) from at least three independent experiments performed in independent preparations. **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way ANOVA with Bonferroni test.

Multiple lines of evidence suggest that PSD-95 and Pyk2 work together to mediate synaptic plasticity through a mechanism involving NMDAR (4143). Moreover, Ca2+ influx through NMDAR was shown to induce Pyk2 clustering at synapses, namely, at the postsynaptic region (43). To determine whether BDNF also induces the synaptic accumulation of Pyk2, hippocampal neurons were stimulated with BDNF and the cells were immunostained for Pyk2, PSD-95, and MAP2 (Fig. 9, A and E to G). Images were analyzed for the number (Fig. 9E) and intensity (Fig. 9F) of synaptic (PSD-95 colocalized; arrowheads in Fig. 9A) Pyk2 puncta per dendritic length (MAP2); the immunoreactivity per synaptic puncta was also analyzed (Fig. 9G). The results were similar for the three parameters analyzed, showing a specific increase in synaptic clustering of Pyk2 after stimulation with BDNF (Fig. 9, A and E to G); this effect of the neurotrophin was decreased upon incubation of the cells with cycloheximide. However, when the cells were stimulated with the neurotrophin for 60 min in the presence of cycloheximide, there was still a small but nonsignificant up-regulation in the number of synaptic puncta containing Pyk2, possibly due to a redistribution of preexisting proteins to the synapse. Using the same paradigm of stimulation, we also evaluated the percentage of synaptic Pyk2 (% of Pyk2 with PSD-95 colocalization) (Fig. 9, A and H) and the percentage of synapses containing Pyk2 (% of PSD-95 with Pyk2 colocalization) (Fig. 9, A and I) in hippocampal neurons stimulated with BDNF in the presence or absence of cycloheximide. We found an increase in the percentage of synaptic Pyk2 and in the percentage of synapses containing Pyk2 at 30 and 60 min after BDNF application, and these alterations were abrogated by pretreatment with cycloheximide (Fig. 9, H and I). Together, these results indicate that BDNF-induced up-regulation of dendritic Pyk2 in cultured hippocampal neurons requires protein synthesis.

In a previous microarray analysis of the transcripts that dissociate from the RNA binding protein hnRNP K after incubation of cultured hippocampal neurons with BDNF, we found that (i) Pyk2 mRNA interacts with hnRNP K and (ii) this interaction is weakened after stimulation with the neurotrophin (16). Accordingly, quantitative polymerase chain reaction (qPCR) experiments showed that Pyk2 mRNA was present in hnRNP K immunoprecipitates from cultured hippocampal neurons, and that a decrease in the coimmunoprecipitation of the RNP with Pyk2 transcripts was observed in extracts prepared from hippocampal synaptoneurosomes stimulated with BDNF for 10 min (Fig. 10, A and B). Therefore, we hypothesized that hnRNP K could be involved in BDNF-induced up-regulation of dendritic Pyk2 protein. To test this hypothesis, the effect of BDNF on the dendritic distribution of Pyk2 was analyzed in hippocampal neurons after down-regulation of hnRNP K with a previously validated shRNA (sh6–hnRNP K) (16). A scramble sequence (sh1-Scramble) lacking homology to any known mammalian mRNAs was used as a control. After fixation, neurons were immunostained for Pyk2, GFP (the transduction marker), and MAP2, and the results were analyzed for the number (Fig. 10, C and D) and intensity (Fig. 10, C and E) of Pyk2 puncta per dendritic length (MAP2) and for the Pyk2 immunoreactivity per puncta (Fig. 10, C and F). hnRNP K knockdown (sh6-RNP K) suppressed the effect of BDNF on the dendritic distribution of Pyk2, whereas no effect was observed in hippocampal neurons transduced with sh1-Scramble.

Fig. 10 hnRNP K mediates the effects of BDNF on dendritic Pyk2 protein abundance.

(A) Immunoprecipitation (IP) of and Western blotting (WB) for hnRNP K from cultured hippocampal neuron lysates. Blot is representative of three experiments performed in independent preparations. (B) Amount of Pyk2, GluA1, and NPAS4 mRNA that coimmunoprecipitated with hnRNP K in extracts from cultured hippocampal neurons (DIV 15), stimulated with BDNF (50 ng/ml for 20 min) relative to controls, assessed by qPCR. Data are means ± SEM from three or four independent experiments performed in distinct preparations as indicated (n). *P < 0.05 versus control by Student’s t test. (C to F) Effect of knockdown of hnRNP K on the BDNF-induced increase in dendritic Pyk2 protein abundance in hippocampal neurons. Cultures were infected at DIV 10 with control [sh1-Scramble (sh1-Scrbl)] or hnRNP K–targeted (sh6–hnRNP K) shRNA and, at DIV 14, were either unperturbed or stimulated with BDNF (50 ng/ml for 30 min) and then fixed and immunostained for Pyk2, GFP, and MAP2. Representative images are shown in (C). Images [represented in (C); scale bar, 5 μm] were analyzed for the number (D) and intensity (E) of Pyk2 puncta per dendritic length and for the Pyk2 immunoreactivity per puncta (F). Results are expressed as percentage to sh1-Scramble control. **P < 0.01, ****P < 0.0001 by one-way ANOVA with Bonferroni test. (G to J) As in (C) to (F) assessing the effect of overexpression of hnRNP K compared with treatment with BDNF on dendritic Pyk2 protein abundance in hippocampal neurons transfected at DIV 11 and 12. Data were quantified relative to the GFP-Empty control. Scale bar, 5 μm. (K) Assessment of the total number of GFP–hnRNP K puncta per dendritic length under the same conditions as described in (G) to (J), as a control. Data are means ± SEM from the indicated number of neurons (n) in at least three independent experiments. #P < 0.05, ##P < 0.01, ####P < 0.0001 versus GFP-Empty control by one-way ANOVA followed by the Bonferroni test.

To further investigate the role of hnRNP K in BDNF-induced up-regulation in dendritic Pyk2, we tested the effect of the neurotrophin (applied for 30 min) in hippocampal neurons expressing a GFP-tagged hnRNP K construct (GFP–hnRNP K) or with GFP alone (GFP-Empty). After fixation, neurons were immunostained for Pyk2, GFP (transfection marker), and MAP2 (Fig. 10G). Overexpression of hnRNP K increased the number and intensity of Pyk2 puncta in dendrites, as well as the Pyk2 immunoreactivity per puncta, to similar levels as those observed in hippocampal neurons stimulated with BDNF under control conditions (transfected with GFP) (Fig. 10, G to J). Under the same conditions, there was a significant increase in the number and area of Pyk2 puncta that colocalized with PSD95, whereas the immunoreactivity of synaptic Pyk2 per puncta was not changed upon transfection of GFP–hnRNP K (fig. S2). Furthermore, a substantial fraction of the hnRNP K immunoreactivity was found to colocalize with Pyk2 (fig. S2). After up-regulation of hnRNP K protein levels, BDNF affected neither the number and intensity of Pyk2 puncta along dendrites nor the Pyk2 immunoreactivity per puncta, in contrast with the results obtained under control conditions (GFP-Empty). Control experiments showed no significant effect of BDNF on the total number of GFP–hnRNP K puncta along dendrites under the experimental conditions used (Fig. 10K). Together, these findings show a role for hnRNP K in BDNF-induced dendritic increase in Pyk2 by a mechanism likely involving the control of Pyk2 mRNA nuclear export, its transport, and/or its loading onto polysomes, which ultimately interferes with BDNF-induced local translation of Pyk2 at dendrites.

DISCUSSION

BDNF is an important mediator of the LTP component that depends on protein synthesis, but the molecular mechanisms of action of the neurotrophin are poorly understood. In this work, we show that BDNF up-regulates the synaptic expression and activity of GluN2B-containing NMDAR (sensitive to conantokin G) in cultured hippocampal neurons by inducing the local synthesis of the nonreceptor tyrosine kinase Pyk2. Furthermore, we found that BDNF dissociates Pyk2 transcripts from the RNA binding protein hnRNP K, which may be important to make them available for translation. The observed up-regulation of synaptic GluN2B-containing NMDAR correlates with an enhanced stability of the receptors in this compartment and may partly account for the effect of BDNF on the late phase of LTP, because previous studies showed that an increase in the expression of this subunit is associated with an improvement of synaptic plasticity and memory formation in mice (44, 45). NMDAR-containing GluN2B subunits are characterized by a higher Ca2+ permeability and charge transfer, in addition to slower deactivation, when compared with receptors containing GluN2A subunits (23). The C-terminus of GluN2B subunits interacts with CaMKII, which is important in LTP in the hippocampus. In particular, CaMKII phosphorylates GluA1 AMPA receptor subunits on Ser831, thereby increasing the delivery of the receptors to the membrane (46).

The BDNF-induced up-regulation of the synaptic expression and activity of NMDAR-containing GluN2B subunits was abrogated by shRNAs specific for Pyk2, and the increase in synaptic GluN2B immunoreactivity was also inhibited upon transfection of hippocampal neurons with a dominant negative form of the kinase. Pyk2 also mediated the effects of BDNF on the extrasynaptic distribution of GluN2B and was required for the maintenance of this population of receptors on the dendritic plasma membrane, including the synapse, under resting conditions. These results suggest that the effects of BDNF on the synaptic expression of GluN2B-containing NMDAR may be secondary, at least in part, to the increase in their total surface expression. Accordingly, analysis of the dynamics of single receptors at the synapse using QD imaging showed an increased stability of GluN2B subunits at the synaptic and extrasynaptic compartments in hippocampal neurons stimulated with BDNF.

The effect of BDNF in the up-regulation of surface receptors containing GluN2B subunits is also likely to be mediated by phosphorylation of the receptors on tyrosine residues, which prevents their interaction with the μ2 subunit of the adaptor protein complex AP2 and consequent association with endocytic clathrin-coated vesicles (38, 47). GluN2B subunits are phosphorylated by the Fyn kinase (48, 49), a member of the SFK, which act downstream of Pyk2 (35). Pyk2 and SFK belong to the NMDAR complex (35, 36), and the activation of the latter kinases is thought to occur after dimerization of Pyk2 (50), which triggers the autophosphorylation of the kinase on Tyr402 and recruitment of SFK (51). This is in agreement with the observed increase in Pyk2 phosphorylation on Tyr402 after stimulation of synaptoneurosomes with BDNF, but whether activation of the tyrosine kinase is triggered by an increase in the [Ca2+]i as previously suggested (50) remains to be investigated. Autophosphorylation of Pyk2 after NMDAR activation depends on a direct interaction with PSD-95. The Pyk2-PSD95 interaction also contributes to the accumulation of the kinase at PSDs, thereby contributing to the regulation of NMDAR activity (41, 43). This evidence indicates that Pyk2 is well positioned to regulate NMDAR. Recent studies have suggested that BDNF up-regulates Pyk2 phosphorylation through down-regulation of striatal-enriched protein tyrosine phosphatase 61 (STEP61), a protein phosphatase that dephosphorylates GluN2B subunits and Pyk2 (30, 52, 53). The proteasome-mediated down-regulation of STEP may enhance the effects of BDNF on the synaptic distribution of NMDAR-containing GluN2B subunits.

The effect of BDNF on the surface expression of GluN2B was found to be dependent on protein synthesis, in accordance with the reported role of translation activity in the neurotrophin-evoked up-regulation of NMDAR-mediated mEPSCs (16). A previous study showed that GluN2A, but not GluN2B, is dendritically translated and inserted in the postsynaptic membrane upon synaptic activity in the hippocampus (54). These results suggest that translation activity may be required for the synthesis of an intermediary protein, possibly Pyk2, which can up-regulate GluN2B-containing NMDAR at the synapse, rather than “de novo” synthesis of the receptor subunit. This hypothesis is supported by several observations: (i) Pyk2 was found to be translated in synaptoneurosomes stimulated with BDNF, as determined by analysis of the transcripts present in polyribosomal fractions and by the total Pyk2 protein content; (ii) the synthesis of Pyk2 at the synapse is up-regulated upon stimulation of cultured hippocampal neurons with BDNF, as determined with the FUNCAT-PLA method; (iii) BDNF up-regulated the levels of Pyk2 at the synapse in cultured hippocampal neurons by a mechanism sensitive to the translation inhibitor cycloheximide; (iv) knockdown of Pyk2 and transfection with a dominant negative form of the kinase abolished the effect of BDNF on the synaptic expression of GluN2B; and (v) overexpression of Pyk2 enhanced the surface expression of GluN2B to the same extent as BDNF stimulation. Together, these results show that Pyk2 is a previously unknown target for BDNF-induced local translation, with a substantial impact on the activity of the synapse. In addition to Pyk2, several other synaptic proteins were found to be locally translated in response to BDNF-TrkB signaling, including Arc, Homer2, GluA1, and CaMKII (7, 8, 17, 5558). In the present study, we also observed that the effect of BDNF in increasing the number of synaptic Pyk2 puncta along dendrites was partly insensitive to cycloheximide, suggesting de novo formation of Pyk2 synaptic clusters through recruitment of an extrasynaptic pool of the kinase.

The local synthesis of proteins at the synapse depends on the availability of a selected population of transcripts, which are transported along dendrites in RNA granules (4, 12). Stimulation of cultured hippocampal neurons with BDNF regulates the dendritic traffic of RNA binding proteins (16, 59) and induces the dissociation of P-bodies, a class of RNA granules that were proposed to participate in the control of local protein synthesis in the dendritic compartment (15). We previously showed that the RNA binding protein hnRNP K plays an important role as mediator of the effect of BDNF on dendritic mRNA metabolism. Studies performed in hippocampal neurons showed that BDNF-TrkB signaling induces the release from hnRNP K of a large number of transcripts coding for synaptic proteins and suggested a potentially important role for this RNA binding protein in BDNF-dependent forms of synaptic plasticity. The effect of BDNF in inducing the synaptic delivery of hnRNP K (16) is also expected to further enhance the delivery of Pyk2 transcripts to be used in local translation. Accordingly, in the present work, we observed a decrease in the coimmunoprecipitation of hnRNP K with Pyk2 mRNA after stimulation of synaptoneurosomes with BDNF, which correlated with an increase in the local synthesis and accumulation of the kinase. Furthermore, knockdown of hnRNP K suppressed the effects of BDNF on the synaptic accumulation of Pyk2, as well as on the synaptic expression of GluN2B. Previous studies also showed a role for hnRNP K in BDNF-induced up-regulation of NMDAR-mediated mEPSCs (16). These results point to a key role of hnRNP K in the regulation of local translation of Pyk2 downstream of TrkB receptor activation, which affects the activity of NMDAR.

Together, our results indicate a mechanism in which endogenous BDNF released by synaptic activity triggers local translation of Pyk2, which ultimately increases the accumulation of this protein in PSDs, thereby contributing to the activation of downstream signaling mechanisms coupled to the up-regulation of synaptic GluN2B-containing NMDAR. Activation of this pathway may account, at least in part, for the effects of BDNF on synaptic plasticity because Pyk2 was shown to mediate hippocampal-dependent memory and LTP (35, 60) and up-regulation of GluN2B expression is also associated with an improvement of synaptic plasticity and memory formation in mice (44, 45). In addition to the effects on the synaptic expression of NMDAR, BDNF also induces the local synthesis and delivery of AMPA receptors to the synapse in hippocampal neurons (61), further contributing to the potentiation of glutamatergic synapses. The effects of BDNF on the synaptic proteome are more complex because it also locally regulates the degradation of proteins by the ubiquitin-proteasome system (62, 63) to tightly control the synaptic proteome in this form of plasticity.

MATERIALS AND METHODS

Hippocampal cultures

High-density hippocampal cultures were prepared from the hippocampi of E18-E19 Wistar rat embryos after treatment with trypsin (0.06%; 15-min incubation at 37°C; Gibco–Life Technologies) in Ca2+- and Mg2+-free Hanks’ balanced salt solution (HBSS) (5.36 mМ KCl, 0.44 mM KH2PO4, 137 mM NaCl, 4.16 mM NaHCO3, 0.34 mM Na2HPO4·2H2O, 5 mM glucose, 1 mM sodium pyruvate, 10 mM Hepes, and 0.001% phenol red (pH 7.2)]. The hippocampi were then washed with HBSS containing 10% fetal bovine serum (Gibco–Life Technologies), to stop trypsin activity, and transferred to Neurobasal medium (Gibco–Life Technologies) supplemented with SM1 supplement (1:50 dilution; STEMCELL Technologies), 25 μM glutamate, 0.5 mM glutamine, and gentamicin (0.12 mg/ml) (Gibco–Life Technologies). The cells were dissociated in this solution and then plated in six-well plates (85.5 × 103 cells/cm2) coated with poly-d-lysine (0.1 mg/ml) for biochemical purposes (Western blot and RNA coimmunoprecipitation) or on poly-d-lysine–coated coverslips (80 × 103 cells/cm2) for the analysis of NMDAR mEPSCs. Cultures were maintained in a humidified incubator with an atmosphere of 5% CO2/95% air at 37°C for 14 to 17 days and then stimulated with BDNF (50 ng/ml) (PeproTech) for the indicated periods of time.

Low-density hippocampal cultures were prepared as previously described (64). Briefly, hippocampi were dissected from E18 rat embryos, and the cells were dissociated using trypsin (0.25%) before plating in neuronal plating medium (minimum essential medium supplemented with 10% horse serum, 0.6% glucose, and 1 mM pyruvic acid) at a final density of 0.5 to 2.4 × 103 cells/cm2 on poly-d-lysine–coated glass coverslips. After 2 to 4 hours, coverslips were flipped over an astroglial feeder layer in Neurobasal medium (Invitrogen) supplemented with SM1 supplement (1:50 dilution; STEMCELL Technologies), 25 μM glutamate, 0.5 mM glutamine, and gentamicin (0.12 mg/ml) (Gibco–Life Technologies). The neurons grew face down over the feeder layer but were kept separate from the glia by wax dots on the neuronal side of the coverslips. To prevent overgrowth of glial cells, neuron cultures were treated with 10 μM 5-fluoro-2′-deoxyuridine (Sigma-Aldrich) after 3 DIV. Cultures were maintained in a humidified incubator with an atmosphere of 5% CO2/95% air, at 37°C, for up to 3 weeks, feeding the cells once per week with the Neurobasal medium described above but without glutamate added. At DIV 14 and 15, neurons were stimulated for 30 or 60 min with BDNF (50 ng/ml) (PeproTech). Where indicated, the cells were pretreated for 45 min with the protein synthesis inhibitor cycloheximide (50 μg/ml; Merck-Millipore) or with the vehicle dimethyl sulfoxide (DMSO; 1:1000 dilution; Sigma-Aldrich) as control. After stimulation, cells were fixed and the immunocytochemistry protocol was performed as described in the next section.

Immunocytochemistry and quantitative image analysis

To label surface GluN2B-containing NMDARs, live neurons (low-density hippocampal cultures) were incubated for 15 min at 37°C with an antibody against an extracellular epitope of the GluN2B N terminus (1:100; AGC-003, Alomone Labs) diluted in conditioned culture medium. Neurons were then fixed for 15 min in 4% sucrose and 4% paraformaldehyde in phosphate-buffered saline (PBS) [137 mM NaCl, 2.7 mM KCl, 1.8 mM KH2PO4, 10 mM Na2HPO4·2H2O (pH 7.4)] at room temperature and permeabilized with PBS + 0.3% (v/v) Triton X-100 for 5 min at 4°C. The preparations were then incubated in 10% (w/v) bovine serum albumin (BSA) in PBS for 30 min at 37°C to block nonspecific staining and incubated with the appropriate primary antibody [anti-MAP2 (1:10,000; ab5392, Abcam), anti–PSD-95 (1:200; 7E3-1B8, Thermo Scientific), and anti-vGlut1 (1:5000; AB5905, Millipore)] diluted in 3% (w/v) BSA in PBS (overnight, 4°C). After washing six to seven times in PBS, cells were incubated with the appropriate secondary antibody [Alexa Fluor 488–conjugated anti-mouse (1:1000; A-11001, Thermo Fisher Scientific), Alexa Fluor 488–conjugated anti-rabbit (1:1000; A-11034, Thermo Fisher Scientific), Alexa Fluor 568–conjugated anti-mouse [1:1000; A-11004, Thermo Fisher Scientific], Alexa Fluor 568–conjugated anti-rabbit (1:1000; A-11036, Thermo Fisher Scientific), Alexa Fluor 647–conjugated anti–guinea pig (1:500; A-21450, Thermo Fisher Scientific), and AMCA (aminomethylcoumarin)–conjugated anti-chicken (1:200; #103-155-155, Jackson ImmunoResearch)] diluted in 3% (w/v) BSA in PBS (45 min, 37°C). The coverslips were mounted using fluorescent mounting medium (DAKO). In experiments aiming at labeling only intracellular proteins, neurons were fixed immediately after stimulation as described above and the same procedure was followed.

Fluorescence imaging was performed on a Zeiss AxioObserver Z.1 microscope using a 63× 1.4 numerical aperture (NA) oil objective and a Zeiss Axio Imager.Z.2 microscope using a 63× 1.4 NA oil objective, both equipped with a Zeiss HRm AxioCam. Images were quantified using the Fiji image analysis software. For quantification, sets of cells were cultured and stained simultaneously and imaged using identical settings. The region of interest was randomly selected, and the dendritic length was measured on the basis of MAP2 staining. The protein signals were analyzed after setting the appropriate thresholds, and the recognizable puncta under those conditions were included in the analysis. For each experiment, similar threshold levels were used to quantify the number and the integrated intensity of puncta in dendrites. Measurements were performed in three to six independent preparations, and at least 10 cells per condition were analyzed for each preparation. In the case of transfected and infected cells overexpressing sh1-Scramble, sh5–hnRNP K, sh6–hnRNP K, GFP, GFP–hnRNP K, shA2-Pyk2, shA4-Pyk2, Flag-Empty, Pyk2-Y402F, and Pyk2-WT, only neurons positive for infection/transfection markers were selected.

To analyze GluN2B surface expression in nontransfected cells, the PSD-95 and vGlut1 signals were thresholded and their colocalization was determined. The surface GluN2B signal was measured after thresholds were set so that recognizable puncta were included in the analysis. Surface GluN2B signal present in glutamatergic synapses was obtained by measuring the surface GluN2B puncta positive for both PSD-95 and vGlut1. The results were represented per density of excitatory synapses (number of positive PSD-95–vGlut1 puncta that colocalized per dendritic length). To quantify the surface GluN2B immunoreactivity in transfected cells, digital images were subjected to a user-defined intensity threshold, to select puncta and measured for puncta intensity and number, for the selected region. The synaptic GluN2B puncta were identified by their overlap with thresholded vGlut1 signal. The results were represented per dendritic length.

For quantification of the Pyk2, PSD-95, vGlut1, and hnRNP K signals, digital images were subjected to a user-defined intensity threshold to select puncta, and the immunoreactivity was quantified for puncta intensity and number, for the selected region. The synaptic Pyk2 puncta were selected by their overlap with thresholded PSD-95 signal. The results were represented per dendritic length. To quantify hnRNP K immunoreactivity, digital images were subjected to a user-defined intensity threshold to select puncta and measured for puncta intensity. The results were represented per dendritic length.

Analysis of NMDAR-mediated mEPSCs

Cultured hippocampal neurons (80 × 103 cells/cm2) with pyramidal morphology (DIV 15 to 17) were whole-cell voltage–clamped to −60 mV, at room temperature, in a Mg-free Tyrode’s solution containing 150 mM NaCl, 4 mM KCl, 10 mM glucose, 10 mM Hepes, and 2 mM CaCl2 (pH 7.35) (310 mOsm). To record and isolate NMDAR-mediated mEPSCs, 6-cyano-7-nitroquinoxaline-2,3-dione [CNQX; 10 μM; Tocris (AMPA/kainate receptor antagonist)], bicuculline {10 μM; Tocris [γ-aminobutyric acid type A (GABAA) receptor antagonist]}, tetrodotoxin [TTX; 500 nM; Tocris (blocker of voltage-gated Na+ channels)], and glycine [15 μM; Sigma-Aldrich (co-agonist of NMDARs)] were added to the bath solution (65). The electrode solution had the following composition 115 mM Cs-MeSO3, 20 mM CsCl, 2.5 mM MgCl2, 10 mM Hepes, 0.6 mM EGTA, 4 mM Na2-ATP (adenosine triphosphate), and 0.4 mM Na-GTP (guanosine triphosphate) (pH 7.3) (300 mOsm; Sigma) (66). Where indicated, hippocampal neurons were preincubated with cycloheximide (50 μg/ml) or with vehicle (DMSO) (1:1000 dilution) for 15 min before recording the NMDAR-mediated mEPSCs. When the effect of BDNF was tested, hippocampal neurons were preincubated with the neurotrophin (50 ng/ml) for at least 30 min before recording the NMDAR-mediated mEPSCs. Recording electrodes were made of borosilicate glass capillaries and pulled on a horizontal stage Sutter Instrument P-97 puller (resistances, 3 to 4 megohms). Recordings were made without series resistance compensation. Cells were recorded for a period of 5 min, and the baseline for the analysis of NMDAR-mediated mEPSCs was manually determined as the average current level of silent episodes during a recording. Whole-cell recordings from hippocampal neurons were performed using an Axon CNS, a MultiClamp 700B amplifier, an Axon Digidata 1550 A acquisition board, and pClamp software (version 10.5, Molecular Devices). Signals were filtered at 2.8 Hz and sampled at 25 kHz, and the amplitude of NMDAR-mediated currents was analyzed offline with pClamp software (version 10.5, Molecular Devices).

SPT imaging

QD staining of surface GluN2B was performed as previously described (67, 68), with some modifications. Briefly, rabbit anti-NMDAR 2B (GluN2B) (1:10; AGC-003, Alomone Labs) antibodies were premixed with anti-rabbit QD 655 (1:15; Q11422MP, Life Technologies) in PBS for 30 min, and the solution was supplemented with casein (threefold concentrated, Vector Labs) during the last 10 min of incubation to prevent nonspecific binding. Neurons were then incubated with the diluted antibody-QD premix [1:50, in 148 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 10 mM glucose, and 10 mM Hepes (pH 7.4)] for 3 min at room temperature.

Excitatory synapses were identified by transfecting neurons with PSD95-FingR-GFP, a disulfide-free intrabody that allows visualizing excitatory synapses in real time, in living neurons, with high fidelity and without affecting neuronal function (69). SPT experiments were performed using an inverted widefield microscope (Zeiss AxioObserver Z1 microscope) equipped with a 63× 1.4 NA oil immersion objective and a digital CMOS camera (ORCA-Flash4.0). The microscope was equipped with a large stage incubator, allowing to keep the cells at 37°C during the experiments. QD fluorescence was monitored over time by acquiring 1200 consecutive frames with an acquisition time of 50 ms using the ZEN Blue software 2012 (excitation filter BP 425/50 and emission filter BP 655/30, Chroma).

SPT analysis

Single QDs were recognized by their diffraction-limited fluorescence spot shape and characteristic blinking and were tracked with a 50-ms time resolution. QD spatial coordinates were identified in each frame as sets of >4 connected pixels using a two-dimensional object (67). Continuous tracking between blinks was reconstructed for the particles that presented 2 to 3 pixels (0.32 to 0.48 μm) of maximal position change between two frames and maximal dark periods of 25 frames (1.25 s). MSD curves were calculated for reconnected trajectories of at least 20 frames (67). The QDs were considered synaptic if colocalized with PSD-95 dendritic clusters for at least five frames. Instantaneous diffusion coefficients, D, were calculated from linear fits of the first four values of the MSD versus time plot according to the following equation: MSD(t) = <r2> (t) = 4Dt. The resolution limit for diffusion was 0.0075 μm2/s (67), whereas the resolution precision was ∼40 nm. The diffusive properties of the mobile receptor population were described as their median ± interquartile range (IQR), defined as the interval between the 25th and 75th percentiles. The software used in QD tracking was provided by D. Choquet (Bordeaux, France)

Synaptoneurosome preparation

Synaptoneurosomes were prepared as previously described (17) with slight modifications (16). Briefly, four to six hippocampi were dissected from adult Wistar rats and tissue was minced with scissors and homogenized with a KONTES Dounce tissue grinder in a buffer containing 0.32 M sucrose, 10 mM Hepes-tris (pH 7.4), and 0.1 mM EGTA using, first, a pestle with large clearance, 0.889 to 0.165 mm (8 to 10 strokes), followed by a small clearance pestle, 0.025 to 0.076 mm (8 to 10 strokes). After centrifugation for 3 min at 1000g, the supernatant was collected and passed initially through nylon membranes (150 and 50 μm, VWR) and finally through an 8-μm pore size filter (Millipore). The flow-through was centrifuged for 15 min at 10,000g, and the resulting pellet was resuspended in incubation buffer [8 mM KCl, 3 mM CaCl2, 5 mM Na2HPO4, 2 mM MgCl2, 33 mM tris, 72 mM NaCl, and 100 mM sucrose (pH 7.4)]. All procedures were performed at 4°C. Before stimulation, synaptoneurosomes were prewarmed at 30°C during 5 min in the same buffer. Incubation with BDNF (50 ng/ml; PeproTech) was performed at the same temperature, and a control experiment in the absence of the neurotrophin was also performed for each time point considered. Where indicated, synaptoneurosomes were preincubated with cycloheximide (50 μg/ml; Merck-Millipore) or with the vehicle, DMSO (1:1000 dilution; Sigma-Aldrich), for 15 min at 30°C. Synaptoneurosomes were then briefly centrifuged in a Minispin microcentrifuge (Eppendorf) for 30 s, and the pellet was resuspended in radioimmunoprecipitation assay (RIPA) buffer [150 mM NaCl, 50 mM tris-HCl, 5 mM EGTA, 1% Triton, 0.5% dissolved organic carbon (DOC), and 0.1% SDS (pH 7.5)] supplemented as indicated for the preparation of extracts, followed by sonication and protein quantification using the bicinchoninic acid (BCA) method. For polysome extraction, the pellet was resuspended in mammalian lysis buffer [15 mM tris-HCl (pH 8), 5 mM MgCl2, 0.3 M NaCl, 0.5 mM dithiothreitol (DTT), cycloheximide (0.1 mg/ml) (Merck-Millipore), and 1% Triton X-100] supplemented as indicated for isolation of polysomes.

Preparation of hippocampal culture extracts

Hippocampal cultures with 15 DIV (85.5 × 103 cells/cm2) were washed twice with ice-cold PBS and once more with PBS buffer supplemented with 1 mM DTT and a cocktail of protease inhibitors {0.1 mM phenylmethylsulfonyl fluoride (PMSF) and CLAP [chymostatin (1 μg/ml), leupeptin (1 μg/ml), antipain (1 μg/ml), and pepstatin (1 μg/ml)]; Sigma}. The cells were then lysed with RIPA buffer supplemented with 50 mM sodium fluoride (NaF), 1.5 mM sodium orthovanadate (Na3VO4), and the cocktail of protease inhibitors. After sonication and centrifugation at 16,100g for 10 min at 4°C, protein in the supernatants was quantified using the BCA assay kit (Pierce). Samples were then denaturated with 2× concentrated denaturing buffer [125 mM tris (pH 6.8), 100 mM glycine, 4% SDS, 200 mM DTT, 40% glycerol, 3 mM Na3VO4, and 0.01% bromophenol blue] for 5 min at 95°C, and proteins of interest were analyzed by Western blot.

Western blotting

Samples were resolved by SDS–polyacrylamide gel electrophoresis in 10% polyacrylamide gels. For Western blot analysis, proteins were transferred onto a polyvinylidene difluoride membrane (Millipore) by electroblotting (40 V, overnight at 4°C). The membranes were blocked for 1 hour with 5% skim milk in TBS-T [20 mM tris, 137 mM NaCl (pH 7.6), supplemented with 0.1% Tween 20], and probed with the primary antibody overnight at 4°C [anti–hnRNP K (1:1000; sc-28380, Santa Cruz Biotechnology), anti-Pyk2 (1:500; sc-1514, Santa Cruz Biotechnology), and anti-pPyk2 (1:1000; 44-618G, Invitrogen)]. After several washes with TBS-T, the membranes were incubated with an alkaline phosphatase–conjugated immunoglobulin G (IgG) secondary antibody (anti-mouse or anti-rabbit, depending on the primary antibody host-species; Jackson ImmunoResearch) for 1 hour at room temperature. The membranes were washed again, and immunostaining was visualized by the enhanced chemifluorescence method on the Storm 860 Gel and Blot Imaging System (GE Healthcare). Quantification of the immunoreactivity was performed with the ImageQuant Software under linear exposure conditions.

In situ visualization of newly synthesized proteins

Fluorescent in situ tagging of newly synthesized Pyk2 was performed combining FUNCAT and PLA techniques as previously described (40). Briefly, for the FUNCAT part of the assay, the culture medium was replaced with a methionine-free Hepes-buffered solution (HBS; 116 mM NaCl, 25 mM glucose, 10 mM Hepes, 5.4 mM KCl, 0.8 mM MgSO4, 1 mM NaH2PO4, 1.8 mM CaCl2, and 25 mM NaHCO3) supplemented with SM1, for 30 min, to deplete endogenous methionine. Metabolic labeling was performed in supplemented HBS, containing 4 mM AHA, for 30 min, in the absence (control) or presence of BDNF (50 ng/ml). After metabolic labeling, cells were further incubated in conditioned culture medium for 15 min. After labeling, hippocampal neurons were washed twice with PBS-MC [PBS (pH 7.4), 1 mM MgCl2, and 0.1 mM CaCl2] and fixed for 15 min in PFA-sucrose (4% paraformaldehyde and 4% sucrose in PBS) at room temperature, washed, permeabilized, and blocked in B-Block with Triton X-100 (10% horse serum, 5% sucrose, 2% BSA, and 0.1% Triton X-100) for 1 hour and 30 min at room temperature. The cells were then washed three times (10 min each) with PBS at pH 7.8 before the click reaction.

The CuAAC click reaction was performed overnight at room temperature, using a solution containing 200 μM triazole ligand tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA), 25 μM biotin alkyne tag, 500 μM tris(2-carboxyethyl)phosphine (TCEP), and 200 μM CuSO4 in PBS (pH 7.8). After the click reaction, cells were washed with the FUNCAT-wash buffer (0.5 mM EDTA and 1% Tween 20) three times (10 min each) and twice with PBS (pH 7.4) for 10 min each before processing for PLA.

Detection of newly synthesized proteins by PLA was carried out using anti-biotin (#B7653, Sigma) and anti-Pyk2 antibodies, rb PLAplus and ms PLAminus probes as secondary antibodies, and the “Duolink Detection reagents Red” (Sigma) for ligation, amplification, and label probe binding, according to the manufacturer’s recommendations. Briefly, cells were incubated with the primary antibodies diluted in B-blocking buffer (overnight at 4°C). After washing, PLA probes were applied in a 1:10 dilution for 1 hour at 37°C, washed six times with wash buffer A (0.01 M tris, 0.15 M NaCl, and 0.05% Tween 20), and incubated for 30 min with the ligation reaction prepared according to the manufacturer’s recommendations (Duolink Detection reagents Red, Sigma) in a prewarmed humidified chamber at 37°C. Cells were then washed with wash buffer A, and amplification and label probe binding was performed according to the manufacturer’s recommendations (Duolink Detection reagents Red, Sigma) in a prewarmed humidified chamber at 37°C for 100 min. Amplification was stopped by three washes in 0.2 M tris and 0.1 M NaCl (pH 7.5) followed by washes in PBS (pH 7.4). Cells were postfixed for 10 min at room temperature in PFA-sucrose, washed with PBS, and processed further for immunocytochemistry as described above.

hnRNP K immunoprecipitation and mRNA extraction

High-density cultures of hippocampal neurons (DIV 15) were incubated for 20 min in the presence or absence of BDNF (50 ng/ml) in a humidified incubator with an atmosphere of 5% CO2/95% air at 37°C. The cells were then washed twice with ice-cold PBS before lysis with RIPA buffer [150 mM NaCl, 50 mM tris-HCl (pH 7.4), 5 mM EGTA, 1% Triton, 0.5% DOC and 0.1% SDS (pH 7.5)] supplemented with a cocktail of protease inhibitors {0.1 mM PMSF and CLAP [chymostatin (1 μg/ml), leupeptin (1 μg/ml), antipain (1 μg/ml), and pepstatin (1 μg/ml)]; Sigma-Aldrich}, phosphatase inhibitors (50 mM NaF and 1.5 mM Na3VO4), and ribonuclease (RNase) inhibitor (50 U/ml) (SUPERaseIn, Ambion Applied Biosystems). The extracts were then frozen at −80°C. After thawing, the extracts were centrifuged at 16,100g for 10 min at 4°C to remove a membrane fraction. Protein content in the supernatants was quantified using the BCA assay kit (Pierce, Thermo Fisher Scientific). Antibody-immobilized beads were prepared by incubating 6 μg of anti–hnRNP K or mouse IgG antibodies with 100 μl of Protein G PLUS-Agarose beads (Santa Cruz Biotechnology), overnight at 4°C. The immobilized antibodies were incubated with 1 mg of protein during 1 hour at 4°C, and the beads were washed four times (2 min of centrifugations, 2000g) at 4°C with washing buffer, supplemented as described for the lysis buffer. The RNA contained in pellets resulting from hnRNP K immunoprecipitation was immediately isolated using TRIzol reagent (Invitrogen) following the manufacturer’s specifications. After addition of chloroform and phase separation, the RNA was precipitated by the addition of isopropanol. The precipitated RNA was washed twice with 75% ethanol in RNase-free water, centrifuged, air-dried, and resuspended in 10 to 20 μl of RNase-free water (Gibco, Invitrogen). The RNA concentration was determined using NanoDrop (Thermo Fisher Scientific), and samples were stored at −80°C until further use. Reverse transcription and qPCR analysis of the relative gene expression across the experimental conditions was performed as described below.

Reverse transcription and qPCR

RNAs were transcribed to complementary DNA (cDNA) using a reverse transcription protocol. For first-strand cDNA synthesis, 500 to 1000 ng of isolated RNA were mixed with 4 μl of 5× iScript Reaction Mix, 1 μl of reverse transcriptase, and nuclease-free water, up to a total volume of 20 μl per experimental condition (iScript cDNA Synthesis Kit, Bio-Rad). This kit uses a blend of oligo(dT) and random hexamer primers in the reaction mix. The reaction was performed at 25°C for 5 min, followed by 30 min at 42°C, for primer annealing to the template and cDNA synthesis. The reverse transcriptase was then denatured for 5 min at 85°C, and the sample was cooled to 4°C before storage at −20°C until further use. Equal amounts of RNAs of each condition were used for reverse transcription.

The relative amount of coimmunoprecipitated mRNAs was quantified by qPCR using the SsoFast EvaGreen Supermix (172-5201, Bio-Rad). Two microliters of 1:5 diluted cDNA was used, and the concentration of each primer was 250 nM in a final volume of 20 μl. Primers for quantitative reverse transcription PCR (qRT-PCR) were designed by Beacon Designer 7 software (Premier Biosoft International). The following considerations were taken: (i) GC content of about 50%; (ii) annealing temperature (Ta) of 55 ± 5°C; (iii) secondary structures and primers dimers were avoided; (iv) primer length of 18 to 24 base pairs (bp); (v) final product length of 100 to 200 bp. Primer sequences were as follows: Gria1 (GluA1), ACTACATCCTCGCCAATCTG (forward) and AGTCACTTGTCCTCCATTGC (reverse); PTK2B (Pyk2), GTAGATAGCATTGTGTTAG (forward) and ACTATTGATTAAGCATACTG (reverse); RNA 18S, GCTCCTTACCTGGTTGATCC (forward) and AATTACCACAGTTATCCAAGTAGG (reverse); NPAS4, AATGGAGATATTCAGGCT (forward) and TAGTTATTGGCAGTAATAGG (reverse).

The thermocycling reaction was initiated with activation of Taq DNA polymerase by heating at 95°C during 30 s, followed by 45 cycles of a 10-s denaturation step at 95°C, a 30-s annealing step at the optimal primer temperature of annealing, and a 30-s elongation step at 72°C. The fluorescence was measured after the extension step by the iQ5 Multicolor Real-Time PCR Detection System (Bio-Rad). After the thermocycling reaction, the melting step was performed with slow heating, starting at 55°C and with a rate of 0.5°C per 10 s, up to 95°C, with continuous measurement of fluorescence to allow the detection of nonspecific products.

qPCR data analysis

The comparative threshold cycle (Ct) method was used to quantitate the relative gene expression across the experimental conditions. Ct represents the detectable fluorescence signal above background, resulting from the accumulation of amplified product, and is a proportional measure of the starting target sequence concentration. Ct was measured on the exponential phase, and for every run, Ct was set at the same fluorescence value. Data analysis of the log-transformed expression data was performed using GenEx (MultiD Analysis, Sweden) software for qPCR expression profiling.

Transfection of cultured hippocampal neurons

Constructs were recombinantly expressed in primary cultures of hippocampal neurons at 11 to 12 DIV using the calcium phosphate transfection protocol (70) with minor alterations. Plasmids (2 μg per coverslip) were diluted in tris-EDTA transfection buffer [10 mM tris-HCl and 2.5 mM EDTA (pH 7.3)], and a CaCl2 solution (2.5 M in 10 mM Hepes) was then added, dropwise, to the plasmid DNA–containing solution to give a final concentration of 250 mM CaCl2. This solution was then added to an equivalent volume of Hepes-buffered transfection solution [274 mM NaCl, 10 mM KCl, 1.4 mM Na2HPO4, 11 mM dextrose, and 42 mM Hepes (pH 7.2)], and the mixture was vortexed gently for 2 to 3 s. The precipitated DNA was added dropwise to the coverslips, and the cultures were incubated with the precipitate for 1.5 hours in the presence of kynurenic acid (2 mM). Each coverslip was transferred to a fresh well of the 12-well plate containing 1 ml of conditioned culture medium with kynurenic acid (2 mM) and slightly acidified with HCl (~5 mM final concentration), and the plate was returned to a 5% CO2/95% air incubator at 37°C for 15 min. Coverslips were then transferred to the original dish containing the conditioned medium. The cells were then returned to a 5% CO2/95% air incubator at 37°C to allow expression of the transfected construct. Protein expression was typically for 72 hours, and the experiments were performed at DIV 14 and 15, depending on the day of transfection (DIV 11 and 12).

Plasmids and constructs

The pSHCMV 3XFLAG Pyk2 (Pyk2_WT) and p3XFLAG Pyk2 Y402F plasmids were a gift from J. C. Loftus (Mayo Clinic, Arizona). Lentiviral plasmids sh1_Scramble, sh5_hnRNP K, and sh6_hnRNP K were designed and cloned as described below. Pyk2 knockdown was performed using the sequences ShA2 (ShA2_Pyk2) and ShA4 (ShA4_Pyk2) (71), cloned as described below. In parallel, we also designed and cloned the construct sh1_Scramble. The GFP–hnRNP K (GFP_hnRNP K) plasmid was provided by A. Ostareck-Lederer (Institute of Biochemistry, Martin-Luther-University Halle-Wittenberg, Kurt-Mothes-Strasse), and the GFP_Empty and Flag_Empty plasmids were given by A. L. Carvalho (Center for Neuroscience and Cell Biology, University of Coimbra, Portugal). All plasmid sequences were confirmed by DNA sequencing.

hnRNP K and Pyk2 knockdown in neuronal cultures

TRIP∆U3-E1a-EGFP (pTRIP) lentiviral vectors (72) were used to deliver shRNA for hnRNP K knockdown in neuronal primary cultures, as previously described (16). To obtain shRNA templates, the sense and antisense strands were designed to contain a 19- to 22-nucleotide duplex connected by a short loop structure (5′-TTCAAGAGA-3′) and flanked by 5′-Bgl II and 3′-Hind III restriction sites. The templates used were 5′-gatgaacgctctggatgcg-3′ for the nontargeting (control) shRNA template and 5′-gtaactattcccaaagatt-3′ for hnRNP K, as depicted in table S1. To knock down Pyk2, we used the sequences previously described by Zhang and colleagues (71) and here indicated as ShA2 (ShA2_Pyk2) and ShA4 (ShA4_Pyk2).

After annealing, oligonucleotides were cloned into the Bgl II– and Hind III–digested pSuper (Eco RI) intermediate vector. Then, a fragment containing the H1 promoter and hairpin sequences were obtained from Eco RI–digested pSuper and subcloned into the Eco RI site of the pTRIP lentiviral vector.

Lentivirus construction and transduction of neuronal cultures

Lentiviruses were generated by triple calcium-phosphate transfection of pTRIPshRNA (coding also GFP), pCMV-ΔR8.91, and pMD (VSVG), which encode the VSVG envelope glycoprotein gene and the gag/pol/tat genes, respectively, into human embryonic kidney (HEK) 293 T cells. HEK293T cells were grown for 2 days in 10-cm petri dishes until they reached about 60% confluence. A solution of CaCl2 and DNA [helper plasmids, 10 μg of pCMV-ΔR8.91 and 6 μg of pMD.G(VSVG); plasmid with the specific constructs, 5 μg of pTrip-shRNA] was added dropwise to a solution of 2× HBS [50 mM Hepes, 280 mM NaCl, and 1.5 mM Na2HPO4 (pH 7.0)]. The solution of calcium DNA was added dropwise to 2× HBS and rested for 10 min to form the precipitates. The precipitates were then distributed evenly over the HEK293T cultures. The cells were allowed to incorporate the precipitates for 6 hours and further incubated for about 60 hours to express the plasmid content. During these periods, the cells were maintained at 37°C, with saturating humidity and an atmosphere of 5% CO2/95% air. The supernatant containing viral particles was then collected and concentrated by centrifugation at 60,000 g for 2 hours at 22°C (73). Viral particles were resuspended in 0.1% BSA in PBS and stored at −80°C. Viral titer was calculated as previously described (74).

Neuronal cultures were transduced at DIV 11 with a multiplicity of infection equal to 5, which represents about 80% of neuronal infection. Coverslips with low-density hippocampal neuronal cultures growing over a layer of astroglia cells were transferred to sterile 12-multiwell plates, where the cells were transduced for 6 hours in 500 μl of conditioned media. After that period, the coverslips were gently washed in sterile PBS and then transferred to the wells containing the astroglia cell layer. Neurons were allowed to express the shRNA for 3 days. At 14 DIV, the neurons were processed for immunocytochemistry.

Statistical analysis

The results are presented as means ± SEM. Statistical analysis was performed using parametric tests. Student’s t test was used to compare two groups of samples. Comparisons between multiple groups were performed with one-way ANOVA, followed by Bonferroni’s multiple comparison test, using the Prism 8 software (GraphPad). Statistical tests were interpreted at a 5% significance level, and all the P values presented were two tailed.

SUPPLEMENTARY MATERIALS

stke.sciencemag.org/cgi/content/full/12/586/eaav3577/DC1

Fig. S1. Evaluation of Pyk2 knockdown efficiency by shRNA.

Fig. S2. hnRNP K overexpression increases the synaptic abundance of Pyk2.

Table S1. shRNA sequences.

REFERENCES AND NOTES

Acknowledgments: We thank D. Choquet for providing the software used in QD tracking. Funding: This work was supported by grants from the Portuguese Science and Technology Foundation (FCT), the European Regional Development Fund (ERDF), through the Centro 2020 Regional Operational Programme grants POCI-01-0145-FEDER-028656 (to C.B.D.) and UID/BIM/4501/2013, through the COMPETE 2020–Operational Programme for Competitiveness and Internationalisation, and Portuguese national funds via FCT under projects UID/NEU/04539/2019, PEst-C/SAU/LA0001/2013-2014, PTDC/SAU-NEU/104297/2008, PD/BD/135498/2018 (to P.D.L.), and SFRH/BPD/115546/2016 (to M.M.), and by Liga Portuguesa Contra a Epilepsia. Author contributions: P.A., P.D.L., R.S.C., and M.M. performed experiments and analyzed data. L.C., P.P., and R.D.A. designed the experiments. B.O. analyzed data. C.B.D. designed the project, analyzed data, and wrote the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.
View Abstract

Navigate This Article