Abstract
Mastoparan-7 (Mas-7), an analogue of the peptide mastoparan, which is derived from wasp venom, is a direct activator of Pertussis toxin- (PTX-) sensitive G proteins. Mas-7 produces several biological effects in different cell types; however, little is known about how Mas-7 influences mature hippocampal neurons. We examined the specific role of Mas-7 in the development of dendritic spines, the sites of excitatory synaptic contact that are crucial for synaptic plasticity. We report here that exposure of hippocampal neurons to a low dose of Mas-7 increases dendritic spine density and spine head width in a time-dependent manner. Additionally, Mas-7 enhances postsynaptic density protein-95 (PSD-95) clustering in neurites and activates signaling, increasing the intracellular Ca2+ concentration. To define the role of signaling intermediates, we measured the levels of phosphorylated protein kinase C (PKC), c-Jun N-terminal kinase (JNK), and calcium-calmodulin dependent protein kinase IIα (CaMKIIα) after Mas-7 treatment and determined that CaMKII activation is necessary for the Mas-7-dependent increase in dendritic spine density. Our results demonstrate a critical role for subunit signaling in the regulation of synapse formation.
1. Introduction
G proteins are highly expressed in the mammalian brain and play a critical role in the regulation and development of synaptic transmission because they act as transducers for the G protein-coupled receptors (GPCRs) [1]. They are composed of a guanine nucleotide-binding α subunit () and a complex (). In mammals, 20 different G proteins have been described, each composed of one of the 19α subunits, one of the 5β subunits, and one of the 12γ subunits [2]. In the resting state, is bound to GDP and associated with and a GPCR. This complex is dissociated when binds to GTP, causing the activation of and the complex, allowing them to regulate their downstream effectors [3]. subunits are separated into four families based on sequence homology (, , , ); each of these proteins activates a different pathway [4].
The majority of excitatory synaptic connections in the central nervous system are located on small dendritic protrusions, which are enriched in signaling molecules and serve to compartmentalize individual postsynaptic structures [5]. In mouse and human brains, the G protein subunits , , , , , , , and are present in postsynaptic densities (PSDs) [6], suggesting a key role of these signal transducer proteins in the synaptic regulation.
To study the specific role of G proteins in the regulation of dendritic spines, we used Mas-7, a potent analogue of the peptide mastoparan, which is obtained from the venom of Vespula lewisii [7]. Mas-7 has a substitution of an alanine for a lysine at position 12 [8] and binds to the plasma membrane to form an α-helix structure that activates subunits without requiring the activation of a GPCR [9]. This peptide shows a wide variety of biological effects, including antiviral activity [10], histamine release from mast cells [11], the induction of potent mitochondrial permeability [12], and tumor cell cytotoxicity [13]. However, the effect of Mas-7 in hippocampal neurons has not been studied.
Here, we show that a low dose of Mas-7 activates , causing the switch from GDP to GTP in hippocampal neurons. Functionally, Mas-7 increases dendritic spine density through a Ca2+-dependent mechanism in hippocampal neurons. Mas-7 also activates a variety of Ca2+-sensitive proteins, including CaMKIIα, which is necessary for the increase in dendritic spine density.
These results suggest that G protein activation, especially the subunit activation, may contribute to dendritic spine remodeling in neurons.
2. Materials and Methods
2.1. Reagents
Mas-7 was purchased from Sigma-Aldrich (St. Louis, MO); Fura-2AM from Molecular Probes (Eugene, OR); and KN93 from Calbiochem (San Diego, CA).
2.2. Hippocampal Neuronal Culture
Rat hippocampal cultures were prepared from Sprague-Dawley rats of both sexes at embryonic day 18, as previously described [14]. On day two, neurons were treated with 2 μM cytosine arabinoside for 24 h to avoid glial cell growth. Then, the neurons were cultured with Neurobasal medium supplemented with 1% B27 from Invitrogen (Eugene, OR).
2.3. Measurements of Intracellular Ca2+ in Hippocampal Neurons
Cytosolic Ca2+ signals were determined in cells seeded at 160,000 per 35 mm coverslip; the cells were loaded with 4.5 μM Fura-2-AM for 30 min as previously described [15]. The experiments were performed in an isotonic calcium-free solution (in mM): 140 NaCl, 2.5 KCl, 1.7 MgCl2, 5 glucose, 0.5 EGTA, and 10 HEPES (305 mOsm/L, pH 7.4 with Tris). An Olympus Spinning Disc IX81 microscope was used in live-imaging experiments recording 1 photo every 5 seconds. The increases in cytosolic Ca2+ are represented by the normalized ratio of the fluorescence emitted at 510 nm after excitation at 340 (which determine the probe bound to Ca2+) and 380 nm (probe not bound to Ca2+) relative to the ratio measured prior to cell stimulation. The integration of the area under the curve was performed with GraphPad Prism5 software (La Jolla, CA) using the first minute before the stimuli application as a baseline.
2.4. Immunoprecipitation of Activated Subunit
The activation assay kit from New East Bioscience (#80901, King of Prussia, PA) was used. The protocol recommended by the manufacturer was employed with modifications. Briefly, neurons at 14 days in vitro (DIV) (seeded at 900,000 cells/well) were treated with 1 μM Mas-7 for 5 or 30 min. Then, the cells were lysed with 0.5 mL 1x kit buffer (#30303) and centrifuged at 12,000 ×g 4°C for 10 min. The supernatants were incubated with 1 μL mouse monoclonal antibody specific for bound to GTP (active form) (#26907) and 20 μL A/G agarose beads (#30301) for 2 h at 4°C with orbital rotation. As a positive control, untreated neurons were lysed and then incubated with 10 mM GTPγS (#30302) and 10 mM MgCl2 for 90 min at RT, and as a negative control, the lysed neurons were incubated with 10 mM GDP (#30304) and 10 mM MgCl2. Later, the lysates were washed 3 times and the beads were suspended in 20 μL Laemmli 2x loading buffer and boiled for 5 min. The total level of was detected by immunoblotting with a polyclonal anti- antibody (#21015, 1 : 1000).
2.5. Western Blot
Neurons at 14 DIV were seeded at 400,000 cells/well and treated with 1 μM Mas-7 and were lysed on ice and immediately processed. Immunoblotting was performed as described [16]. The primary antibodies used included mouse anti-CaMKIIα (sc-5306, 1 : 1000), mouse anti-phospho-Tyr286-CaMKIIα (sc-32289, 1 : 1000), rabbit anti-PKCβII (sc-210, 1 : 1000), rabbit anti-β-tubulin (sc-9104, 1 : 1000), mouse anti-GAPDH (sc-32233, 1 : 5000), and rabbit anti-GSK-3β (sc-9166, 1 : 1000) from Santa Cruz Biotechnology Inc. (Santa Cruz, CA); rabbit anti-JNK (#9252, 1 : 1000), rabbit anti-phospho-Thr183/Tyr185-JNK (#4668, 1 : 1000), and rabbit anti-phospho-Ser9-GSK-3β (#9336, 1 : 1000) from Cell Signaling Technology (Beverly, MA); rabbit anti-phospho-Ser660-PKCβII (ab75837, 1 : 10000) and rabbit anti- (ab136535, 1 : 5000) from Abcam (Cambridge, MA); mouse anti-PSD-95 (k28/43, 1 : 1000) from UC Davis/NIH NeuroMab Facility; and mouse anti-β-actin (11978, 1 : 10000) from Sigma (St. Louis, MO). Equal amounts of protein were loaded (20 μg).
2.6. PSD-95 Immunofluorescence and Image Analysis
Neurons at 14 DIV were plated at 35,000 cells/coverslip and treated with 1 μM Mas-7, fixed with a freshly prepared solution of 4% paraformaldehyde plus 4% sucrose in PBS for 20 min at 4°C, permeabilized with 0.2% Triton X-100 for 5 min at room temperature (RT), and then blocked with 1% BSA in PBS (blocking solution) for 30 min at 37°C. This procedure was followed by an overnight incubation at 4°C with anti-PSD-95 (k28/43, 1 : 400) from UC Davis/NIH NeuroMab Facility and Synapsin I (SynI) (sc-20780, 1 : 2000) from Santa Cruz Biotechnology Inc. The neurons were washed with PBS and incubated for 30 min at 37°C with Phalloidin-Alexa-633 and the secondary antibody (Molecular Probes). Images were acquired from 10 microscope fields for each condition with an Olympus Fluoview FV 1000 confocal microscope. To quantify PSD-95 clusters, we used a previously described protocol [17] using NIH ImageJ software (NIH, Baltimore, MD). The synaptic contacts were measured as previously reported [18].
2.7. Transfection and Dendritic Spine Morphology Analysis
Hippocampal neurons plated at 60,000 per poly-D-lysine-coated 12 mm glass coverslip were transfected at 10 DIV with an EGFP plasmid (pEGFP-N1, Clontech, Mountain View, CA) using a NeuroMag kit (KC 30800) from OZ Bioscience (Marseille, France) as described previously [19]. First, the neurons were washed for 30 min with Neurobasal medium. Then, 0.8 μg DNA/1.25 μL magnetobeads per cover were mixed and incubated for 15 min in 100 μL Neurobasal medium at RT. Next, the mix was added by drops to the neurons, and the magnetobeads were allowed to enter the cells by use of the magnet for 15 min (37°C, 5% CO2). Subsequently, the magnet was removed for 40 min, and, finally, the transfected medium was replaced with fresh medium. At 14 DIV, the neurons were depleted for 2 h with Neurobasal medium without B7 supplement before being treated with 1 μM Mas-7 or 10 μM KN93 plus Mas-7 at different times. An Olympus Fluoview FV 1000 confocal microscope was used to obtain digital confocal stacks from 15 to 20 serial images with a Z step size of 0.25 μm. Dendritic Z-stacks were reconstructed using the super-pass module of Imaris software. Accurate reconstruction of spine head diameter was achieved using the approximate circle algorithm with a threshold of 0.8. Ten neurons were analyzed for each condition. The mean spines length and spines head width of each neurite were reported. Spine density was calculated by measuring the total number of spines per neurite length (spine density/10 μm).
2.8. Live-Cell Imaging of Dendritic Spine Morphogenesis
Hippocampal neurons cultured in round 35 mm coverslips at a density of 160,000 cells/coverslip were transfected with EGFP at 11 DIV. Then, at 14 DIV the neurons were placed in the imaging chamber in an isotonic solution (in mM: 1.2 CaCl2, 140 NaCl, 2.5 KCl, 0.5 MgCl2, 5 glucose, and 10 HEPES (305 mOsm/L, pH 7.4 with Tris)). The EGFP-positive neurons were imaged with an Olympus Spinning Disc IX81 microscope every 5 min for 45 min after the treatment with 1 μM Mas-7. The images were processed and analyzed using ImageJ software.
2.9. Statistical Analysis
Statistical analysis was performed using Prism 5 software. The values are expressed as the mean ± standard error of the mean. The statistical significance of differences was assessed with one-way ANOVA with Bonferroni’s posttest for multiple comparisons and with Student’s -test for comparisons between two conditions ( was considered significant). The number of independent experiments is indicated in the corresponding figure legends.
3. Results
3.1. Mas-7 Activates the Subunit in Hippocampal Neurons
To determine whether the Mas-7 peptide activates G proteins, specifically the subunit in cultured rat hippocampal neurons, we treated them with Mas-7 and then performed immunoprecipitation assays, using a commercial specific antibody that recognizes bound to GTP (-GTP), the active form of the protein, in combination with SDS-PAGE and immunoblotting. Additionally, we immunoprecipitated -GTP from lysates of hippocampal neurons that were incubated with nonhydrolyzable GTP (GTPγS) or with GDP as a positive and negative control, respectively.
Our data show that Mas-7 induced activation of the subunit in cultured hippocampal neurons after 5 min of exposure (Figure 1(a), left panel); the activation then declined at 30 min. By contrast, Mas-7 was unable to increase the total protein level, even after 2 h of treatment (Figure 1(b)). Furthermore, incubation with GTPγS produced an increase in the activation of , and GDP incubation decreased the activation (Figure 1(a), right panel). These findings show that Mas-7 produces rapid activation of subunit, suggesting that the -dependent signaling cascade is also activated.
(a)
(b)
(c)
(d)
(e)
3.2. Mas-7 Increases the Intracellular Calcium Concentration in Hippocampal Neurons
In different cell types, mastoparan and Mas-7 produce an increase in intracellular calcium (Ca2+) [20, 21]. For example, in rat cerebellar granule neurons, 15 μM mastoparan produces a robust elevation in intracellular Ca2+ [22]. To assess whether Mas-7 generates a similar effect in cultured hippocampal neurons, we treated them with Mas-7 at a lower dose (1–5 μM) to avoid a toxic effect. To quantify the response of individual neurons to Mas-7, we used Fura-2 AM in a Ca2+-free solution and measured the fluorescence emitted by the probe at 510 nm after the 340/380 excitation in a live imaging experiments. In Figure 1(c) (Fura-2 AM 340/380 ratio images), it is observed that Mas-7 increased the Ca2+ concentration in the soma and dendrites of the hippocampal neurons. Additionally, Mas-7 induced a large elevation of Ca2+ in a concentration-dependent manner (Figure 1(d)). There was a significant difference between the Ca2+ elevations induced by 1 and 5 μM (Figure 1(e)). This finding suggests that the activation of by Mas-7 produces the release of Ca2+ from internal cellular stores in hippocampal neurons.
3.3. Mas-7 Activates CamKII, PKC, and JNK in Mature Hippocampal Neurons
Because Mas-7 increased the Ca2+ concentration in hippocampal neurons, we evaluated whether Mas-7 could activate Ca2+-dependent kinases, such as CaMKIIα and PKCβII. Previous reports in neuronal and nonneuronal cells have shown that mastoparan can induce a rise in Ca2+ via a phospholipase C- (PLC-) dependent mechanism [7, 22]. The ability of Mas-7 to activate these kinases was determined using anti-phospho-Tyr286-CaMKIIα or anti-phospho-Ser-660-PKCβII antibodies in mature hippocampal neurons.
As illustrated in Figure 2, Mas-7 stimulated the activity of CaMKIIα in a time-dependent manner. We found that hippocampal neurons treated with Mas-7 showed a slight but significant increase in the active form of CaMKIIα after 5 min. Then, the phosphorylation of CaMKIIα decreased and remained at a similar level after 1-2 h with respect to the control condition. In addition, Mas-7 treatment increased PKC phosphorylation after 5 min, with a peak at 60 min (Figure 2). The total levels of PKC were not affected by Mas-7 treatment. Moreover, we analyzed the effect of Mas-7 on the phosphorylation state of other kinases, such as JNK and glycogen synthase kinase-3β (GSK-3β). Mas-7 also induced an increase in JNK phosphorylation (p-JNK-Thr183/Tyr185) after 15 min, mainly of the JNK1 isoform, which corresponds to the lower band. However, hippocampal neurons exposed to Mas-7 did not show any change in the phosphorylation of GSK-3β at serine 9 (p-GSK-3β-Ser9), even after 2 h of treatment. These findings indicate that activation by Mas-7 promotes the activation of CaMKII, PKC, and JNK in hippocampal neurons.
3.4. New PSD-95 Clusters Are Induced by Mas-7 Treatment in Hippocampal Neurons
PSD-95 is a scaffold protein that plays a key role in synapse organization, during dendritic spine formation [23]. Here, we examined whether the Mas-7 peptide could regulate the postsynaptic region in mature hippocampal neurons. Specifically, we analyzed whether Mas-7 would induce an increase in PSD-95 clustering in 14 DIV neurons by measuring PSD-95 density and the area of the clusters by immunofluorescence, as described previously [17].
The effect of Mas-7 in the clustering of PSD-95 was evaluated, and a time-dependent increase in the number of PSD-95 clusters was observed. As indicated in Figures 3(a) and 3(b), PSD-95 clustering increased approximately 50% after 60 min of Mas-7 treatment and remained elevated at 2 h in comparison with the control. To evaluate whether the increase in PSD-95 density was due to an increase in the expression of PSD-95, we measured the area of PSD-95 clusters and the levels of PSD-95 protein in total extracts. Mas-7 did not change the area of the PSD-95 clusters (Figure 3(c)) or the PSD-95 total protein level (Figure 3(d)) over the same temporal course for which we observed the increase in the number of PSD-95 clusters, suggesting redistribution of existing PSD-95 proteins, rather than changes in expression.
(a)
(b)
(c)
(d)
Thus, our findings suggest that activation by Mas-7 triggers PSD-95 remodeling that promotes clustering and additional postsynaptic assembly, without changing the expression of the PSD-95 protein.
3.5. Mas-7 Changes the Morphology and Density of Dendritic Spines in Hippocampal Neurons
The presence of PSD-95 clusters in excitatory neurons is well correlated with the number of mature dendritic spines [24]. For this reason, we attempted to determine the role of Mas-7 in dendritic spine formation by transfecting mature hippocampal neurons with EGFP at 10 DIV and then exposed them to 1 μM Mas-7 at different times at 14 DIV. Dendritic spine protrusions below 3 μm in length were analyzed using Imaris software to measure spine length, width, and density. Hippocampal neurons exposed to Mas-7 exhibited increased dendritic spine density after 30 min and 1 h (Figure 4(a)), with a peak at 2 h. Additionally, Mas-7 increased spine head width after 30 min (Figure 4(c)); however, the length of the spines was not significantly affected (Figure 4(d)). An increase in dendritic spine head width has been related to the strength of synaptic transmission [25], which suggests that Mas-7 might also regulate that process.
(a)
(b)
(c)
(d)
(e)
To provide additional evidence of the effect of Mas-7 in the development of dendritic spine protrusions, live cell time-lapse imaging of the formation of dendritic spines was performed. EGFP-transfected neurons were treated for 45 min with Mas-7 (Figure 4(e)), and we determined that Mas-7 produced de novo formation of a dendritic spine. This new protrusion presented a recognizable head and appeared after 30 min of treatment.
Together, these results suggest that the activation of has a regulatory effect on spine morphogenesis in hippocampal neurons.
3.6. The Activation of CaMKIIα Is Involved in the Dendritic Spine Density Increase Induced by Mas-7
Mas-7 is capable of activating CaMKIIα very rapidly; thus, we sought to establish whether this activation was required to produce the increase in dendritic spines density to understand the cellular mechanism of Mas-7. Interestingly, CaMKIIα has been related to dendritic spine formation and regulation [26].
We used KN93, a classic inhibitor of CaMKII activation [27]. In hippocampal neurons, KN93 completely blocked the increase of dendritic spine density induced by Mas-7 at 2 h, without affecting density when applied alone (Figures 5(a) and 5(b)). These results indicate that the activation of CAMKIIα by the increase in intracellular Ca2+ concentration is required for the regulation of spinogenesis induced by Mas-7.
(a)
(b)
3.7. Mas-7 Induces Synaptic Contacts in Hippocampal Neurons
Our previous findings of Mas-7 regulation of spine formation as well as of PSD-95 cluster remodelling (Figures 3 and 4) led us to suggest that Mas-7 is a regulator of the synapse. To address this alternative, hippocampal neurons at 14 DIV were incubated for 1 or 2 h with Mas-7. Treatment with 1 μM Mas-7 significantly increased the number of synaptic contacts after 1-2 h of treatment (Figure 6). The synaptic contacts were observed by the staining for PSD-95 (green) and for the presynaptic protein Syn I (red), where both stains are facing directly. Interestingly, our results suggest that Mas-7 rapidly increases the number of PSD-95 puncta at 1 and 2 h and simultaneously increases the number of synaptic contacts.
(a)
(b)
4. Discussion
In the present work, we studied the effect of Mas-7, a peptide used to pharmacologically activate G proteins, on synaptic structure.
First, our findings show that the subunit is activated in hippocampal neurons by Mas-7 treatment, which is consistent with previous data obtained in other cellular contexts, through a mechanism similar to the action of GPCRs [9, 28]. The rapid activation of occurred after 5 min of Mas-7 exposure in hippocampal neurons, as its analogue mastoparan, because it promotes the dissociation of GDP and enhances the GTP binding [29]. Additionally, mastoparan increases the intrinsic GTPase activity of G proteins, which suggests that the active state of is transient [30]. We observed that, after 30 min of treatment, the activation of decayed, which is consistent with an increase in GTPase activity.
Furthermore, it is possible that Mas-7 can activate other G proteins subunits in hippocampal neurons, such as , because mastoparan has been shown to be able to activate both these subunits in a biochemical assay [9].
In hippocampal neurons, Mas-7 presents an immediate effect, increasing the intracellular Ca2+ in a concentration-dependent manner, as previously reported in neuroblastoma cells [21] and in neutrophils [20]. This increase remained at least 4-5 min after treatment with Mas-7. It is known that, in addition to being an activator of G proteins, Mas-7 is an inhibitor of ATPase activity from the endoplasmic reticulum [31], which could explain the sustained rise. This elevation in the levels of Ca2+ led to the activation of CaMKIIα and PKCβII, two Ca2+-dependent kinases, as well as JNK. The activation of CaMKII was fast, but Mas-7 activated PKC after 1 h. It would be interesting to study how they regulate the biological effects of Mas-7.
On the other hand, it has been suggested that mastoparan treatment can produce apoptosis in cerebellar granule cells through Ca2+ release, probably via activation of a transduction pathway involving PLC and IP3, using 10–20 μM concentration [22]. In the present study, Mas-7 mimicked the increase in Ca2+ from internal stores; however, we used a lower concentration of Mas-7 for a shorter period of time, which did not replicate the apoptotic effect in hippocampal neurons. Higher concentrations of Mas-7 can probably also trigger a similar apoptotic effect in mature hippocampal neurons.
Previous studies have established that is highly expressed in the central and peripheral nervous systems, where it represents approximately 1% of membrane proteins [32]. Additionally, the subunit has been linked to cognitive and memory functions in the adult brain. The corresponding knockout mice exhibit neurological impairments, such as reduced motor control, hyperactivity, hyperalgesia, and a shortened lifespan [33]. Moreover, is required for the formation of associative memory in mushroom body neurons in D. melanogaster [34].
Functionally, we observed that Mas-7 was able to modulate the postsynaptic region in the mammalian CNS. In the postsynaptic region, the regulation of dendritic spines is a key process in neuronal plasticity and memory. Dendritic spines undergo structural modifications in response to a diverse range of stimuli [35]. Here, we showed that Mas-7 can regulate dendritic spine formation, increase dendritic spine density and head width, and increase PSD-95 clustering. Although there are several studies that support a model in which PSD-95 is recruited in an activity-dependent manner to new spines, where it contributes to the stabilization of nascent spines [36, 37], we sought to study whether these new spines could form synapses. We found that Mas-7 also increased the number of synaptic contacts, which suggests that activation is able to generate functional synapses. Further studies are required to demonstrate whether the increase in the dendritic spine density induced by Mas-7 has a positive impact on memory and learning in vivo.
Effects of Mas-7 treatment on other aspects of neuronal development have been previously demonstrated, such as a significant increase in axonal growth in hippocampal neurons [38], which suggests that activation of the subunit generates several structural effects that could produce cytoskeleton remodeling. In particular, we propose that the Ca2+ increase produced by Mas-7 and the subsequent activation of the downstream Ca2+-sensitive kinases PKCβII and CaMKIIα as well as JNK can explain the dendritic spine remodeling. These kinases are known regulators of dendritic spine morphology [26, 39, 40], but, particularly, it is known that CaMKIIα is highly expressed in spines and is important for long-term potentiation (LTP) [41]. We found that blocking the activity of CaMKIIα prevents the Mas-7-dependent increase in dendritic spine density, helping to elucidate the mechanism by which Mas-7 acts. Certainly, understanding the role of PKC and JNK could provide us with a global vision of the involvement of signaling in spine remodeling and synapse formation.
All of these findings indicate that might be important for the regulation and maintenance of synapses. We suggest a mechanism, in which the activation of this G protein subunit increases the levels of Ca2+, activates CaMKIIα to remodel the postsynaptic region, and ultimately leads to the formation of synaptic contacts.
5. Conclusion
In this work, we demonstrated that the peptide Mas-7 produced several biological effects in mature hippocampal neurons, including activation of signaling and of CaMKIIα, JNK, and PKCβII. Functionally, our results suggest that Mas-7 causes dendritic spine remodeling, increases the number of spines, and recruits PSD-95 protein into spines to produce functional synapses.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This work was supported by grants from FONDECYT (no. 1120156) and the Basal Center of Excellence in Science and Technology (CONICYT-PFB12/2007) to Nibaldo C. Inestrosa, a predoctoral grant to Valerie T. Ramírez, and a postdoctoral fellowship from CONICYT (Fondecyt Postdoctorado No. 3140355) to Eva Ramos-Fernández. The authors thank Gloria Méndez for the neuronal culture preparations.