SEARCH

SEARCH BY CITATION

Keywords:

  • calcium;
  • hyperforin;
  • neuritic outgrowth;
  • neuritic outgrowth;
  • synaptic plasticity;
  • TRPC6 channels

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References
Thumbnail image of graphical abstract

The non-selective cationic transient receptor canonical 6 (TRPC6) channels are involved in synaptic plasticity changes ranging from dendritic growth, spine morphology changes and increase in excitatory synapses. We previously showed that the TRPC6 activator hyperforin, the active antidepressant component of St. John's wort, induces neuritic outgrowth and spine morphology changes in PC12 cells and hippocampal CA1 neurons. However, the signaling cascade that transmits the hyperforin-induced transient rise in intracellular calcium into neuritic outgrowth is not yet fully understood. Several signaling pathways are involved in calcium transient-mediated changes in synaptic plasticity, ranging from calmodulin-mediated Ras-induced signaling cascades comprising the mitogen-activated protein kinase, PI3K signal transduction pathways as well as Ca2+/calmodulin-dependent protein kinase II (CAMKII) and CAMKIV. We show that several mechanisms are involved in TRPC6-mediated synaptic plasticity changes in PC12 cells and primary hippocampal neurons. Influx of calcium via TRPC6 channels activates different pathways including Ras/mitogen-activated protein kinase/extracellular signal-regulated kinases, phosphatidylinositide 3-kinase/protein kinase B, and CAMKIV in both cell types, leading to cAMP-response element binding protein phosphorylation. These findings are interesting not only in terms of the downstream targets of TRPC6 channels but also because of their potential to facilitate further understanding of St. John's wort extract-mediated antidepressant activity.

Alterations in synaptic plasticity are considered to play an important role in the pathogenesis of depression. Beside several other proteins, TRPC6 channels regulate synaptic plasticity. This study demonstrates that different pathways including Ras/MEK/ERK, PI3K/Akt, and CAMKIV are involved in the improvement of synaptic plasticity by the TRPC6 activator hyperforin, the antidepressant active constituent of St. John's wort extract.

Abbreviations used
BDNF

brain derived neurotrophic factor

CREB

cAMP-response element binding protein

NGF

nerve growth factor

PBS

phosphate-buffered saline

TRPC

transient receptor canonical

Several ion channels are involved in synaptic calcium influx such as NMDA receptors or voltage dependent calcium channels (Cullen and Lockyer 2002; Cullen 2006). Recently, the transient receptor canonical (TRPC) subfamily was also shown to play an important role in synaptic calcium elevation (Greka et al. 2003; Tai et al. 2008; Zhou et al. 2008; El Hassar et al. 2011). The TRPC channel subfamily comprises seven members, TRPC1 – TRPC7, and belongs to the TRP channel superfamily of non-selective cation channels (Venkatachalam and Montell 2007). In addition to TRPC3, TRPC1 and TRPC5, TRPC6 channels participate in the regulation of calcium influx in the CNS and thereby improve synaptic plasticity ranging from growth cone guidance (Li et al. 2005), spine morphology changes, dendritic outgrowth (Tai et al. 2008; Zhou et al. 2008; Leuner et al. 2013) to cell survival (Du et al. 2010; Lin et al. 2013). We recently described that the TRPC6 activator hyperforin induces neurite outgrowth in PC12 cells and spine morphology changes in CA1 and CA3 hippocampal neurons (Leuner et al. 2007, 2013). Hyperforin is the anti-depressant active constituent of St. Johns wort which is used to treat mild-to-moderate depression (Muller 2003; Leuner et al. 2010). Hyperforin induces not only Na+ influx but also Ca2+ influx in PC12 cells, cortical, and hippocampal neurons (Treiber et al. 2005; Leuner et al. 2007, 2010, 2013; Gibon et al. 2010). In PC12 cells, hyperforin induced TRPC6-dependent Ca2+ elevation mimics the neurotrophic effects of the nerve growth factor (NGF) such as cell differentiation(Leuner et al. 2007). In addition, hyperforin acts as a brain derived neurotrophic factor (BDNF) mimetic in hippocampal neurons modifying dendritic spine morphology via TRPC6 channels (Leuner et al. 2007, 2010, 2013), which resembles the actions of BDNF through the activation of TRPC3 channels. However, the signaling cascade activated by hyperforin induced TRPC6 mediated calcium influx is not finally understood.

Intracellular calcium binding proteins such as calmodulin decode the information of calcium transients into biochemical and cellular changes via the activation of several signaling cascades ranging from mitogen-activated protein kinase (MAPK) pathway, PI3K cascade to the phosphorylation of CAMK II and IV [ca2+/calmodulin-dependent protein kinase (CAMK)] (Agell et al. 2002; Cullen and Lockyer 2002; Cullen 2006). A positive modulation of Ras signaling pathways by calcium is well-reported for neurons and PC12 cells (Sakai et al. 1999; Agell et al. 2002). A sustained activation of the MAPK pathway including Ras/Raf/MEK/ERK induces cell cycle arrest linked to survival, differentiation, and the formation of dendrites (Koh et al. 2002; Alonso et al. 2004). Other effectors of Ras such as the lipid kinase PI3K are also involved in cell survival, neurite outgrowth as well as dendrite formation and regulation of spine growth in hippocampal neurons (Kumar et al. 2005). In addition, enhanced intracellular calcium leads to an activation of CAMKII or CAMKIV which are also involved in axon and dendrite elongation (Wayman et al. 2008). All these kinases have a common phosphorylation target, the nuclear transcription factor cAMP-response element binding protein (CREB) (Lonze and Ginty 2002; Vaudry et al. 2002). By inducing effector genes p-CREB stabilizes synaptic plasticity. Important targets are genes for neurotrophins such as BDNF, which are key regulators of synaptic plasticity including dendrite outgrowth and spine morphology changes (Santos et al. 2010). Importantly, the described signaling cascades are also involved in several disorders of central nervous system such as depression (Pittenger and Duman 2008).

We focused on the above mentioned key proteins, which are known to be activated by calcium and are essential for neurite outgrowth in PC12 cells as well as primary hippocampal neurons: Ras, phosphatidylinositide 3-kinase/protein kinase B (P13K/Akt), mitogen-activated protein kinase/extracellular signal-regulated kinases (MEK/ERK), CAMKII, and CAMKIV and the downstream target, the transcription factor CREB (Vaudry et al. 2002). Interestingly, several signaling cascades are involved in hyperforin mediated neuritic outgrowth. Here, we show that these signaling cascades are involved in hyperforin-mediated neurite outgrowth.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References

Sources and preparation of reagents

Hyperforin was kindly supplied by the preclinical research department of Dr. Willmar Schwabe (Karlsruhe, Germany). NGF (Merck, Darmstadt, Germany) was dissolved in distilled water and prepared in 50 μg/mL stock solution. Other chemicals were dissolved in dimethylsulfoxide and stock solutions were prepared which were diluted before use. For pharmacological treatments, chemicals were pre-incubated for 1 h and present throughout the experiment. Standard laboratory chemicals were obtained from Sigma-Aldrich (Munich, Germany).

Cell culture and transfection of PC12 cells

PC12 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen; Karlsruhe, Germany) supplemented with 10% heat-inactivated fetal calf serum and 5% heat-inactivated horse serum, 50 U/mL penicillin, and 50 μg/mL streptomycin (Sigma-Aldrich) at 37°C in a humidified incubator containing 5% CO2. For transient transfection of PC12 cells a dominant negative TRPC6-YFP (dnTRPC6-YFP) plasmid was used (Hofmann et al. 2002). Transient transfection was conducted using FuGENE 6 transfection reagent (Roche Diagnostics, Grenzach-Wyhlen, Germany). Cells were plated at a density of 5 × 106 and transiently transfected by addition of a transfection cocktail containing 0.5 to 1 μg of DNA and 2 μL of FuGENE 6 transfection reagent in 30 μL of Opti-MEM medium (Invitrogen). After 48 h, cells were harvested and pellets were stored until use at < −20°C. Depending on the batch, PC12 cells differ slightly in basal levels of neurite outgrowth and in intracellular calcium responses after stimuli such as KCl or hyperforin. However, the qualitative effect regarding neurite outgrowth or intracellular calcium increase remains the same between different batches of PC12 cells.

Neurite outgrowth of PC12 cells

Neurite outgrowth assays were performed according to Leuner et al. 2007. Cells were plated at a density of 105 cells/plate (96 mm, polylysin coated) in 15% serum containing medium overnight. The next day, medium was changed to a medium containing 2% serum and NGF (50 ng/mL) or hyperforin (0.3 μM). The neurite length was examined 3 days after different treatment regimes. After 3 days, PC12 cells were fixed with paraformaldehyde solution (4%) and stained with Mayer's hematoxylin eosin solutions. Ten cells from each staining (n = 1) were arbitrarily analyzed and neurite length was detected by using Nikon NIS Elements AR 2.1 software (Nikon GmbH, Düsseldorf, Germany).

Primary culture of dissociated post-natal neurons

Hippocampal neurons were prepared according to Amaral and Pozzo-Miller (2007) (Amaral and Pozzo-Miller 2007). Briefly, hippocampi were dissected from anesthetized P2 rat pups and dissociated with papain (Worthington, Lakewood, NJ, USA). For the preparation of primary hippocampal neurons female time-mated Sprague-Dawley rats were used. Animals were purchased from Charles River Laboratories, Sulzfeld, Germany or Janvier SAS, St. Berthevin, France. All animal care and experimental procedures were in concordance with the German law on animal care and handling of animals and the guidance of the European Community Council Directive. The tissue was triturated to obtain single cells, which were re-suspended in Neurobasal medium containing B-27 supplement, 10 U/mL, penicillin-streptomycin, and l-glutamine. Dissociated cells were plated on petri dishes coated with poly-l-lysine and laminin and cultured for 14 days at 37°C in 5% CO2, 98% relative humidity. Half of the culture medium was changed every 4 days. At day 14, hippocampal neurons were stimulated with hyperforin. For synapsin staining of bona fide synapses, mature hippocampal neurons (14 div) from three different preparations were incubated with 0.3 μM hyperforin or 50 ng/mL BDNF for 48 h, fixed with paraformaldehyde, incubated overnight with a primary antibody against synapsin (1 : 1000, polyclonal guinea pig antibody; Synaptic Systems, Göttingen, Germany) or PSD-95 (1 : 100, polyclonal rabbit antibody; Abcam, Cambridge, UK) followed by an secondary antibody coupled to Alexa-Fluor 640 nm (synapsin) or Alexa-Fluor 488 nm (PSD-95) for 1 h (Invitrogen), and imaged by confocal microscopy (Leica SP5; 63X 1.4 NA objective, Leica microsystems, Wetzlar, Germany). Synapsin and PSD-95 puncta were counted using ImageJ (W.S. Rasband, U.S. National Institutes of Health, Bethesda, Maryland, USA; http://rsb.info.nih.gov/ij/, 1997–2009). In total, 3000–4000 μm of dendrites were analyzed per coverslip.

Western blot analysis

PC12 cells (5 × 106) and primary hippocampal neurons (5 × 105) were stimulated with hyperforin (10 μM) or NGF (50 ng/mL) for 20 min. Cells were harvested with lysis buffer [1 mM EDTA, 0,5% Triton x-100, 5 mM sodium fluoride (NaF), 6 M urea, 10 μg/mL Leupeptin, 10 μg/mL, Pepstatin, 100 μM phenylmethylsulfonyl fluoride, 3 μg/mL Aprotinin, 2.5 mM Na4P2O7, 1 mM Na3VO4 in phosphate-buffered saline (PBS), pH 7.2–7.4]. Cell lysates (35 μg protein) were used for western blotting. After electrophoresis, proteins were transferred on polyvinylidene difluoride membrane (blocking solution 5% milk) and the membrane was incubated with specific antibodies against p-CREB (1 : 200; Santa Cruz Biotechnology, Heidelberg, Germany), total-CREB (1 : 300; Santa Cruz Biotechnology) and p-CAMKIV (1 : 300; Santa Cruz Biotechnology), p-ERK1/2 (1 : 1000; Cell Signaling, Frankfurt, Germany), total-ERK1/2 (1 : 1000; Cell Signaling), p-MEK1/2 (1 : 1000; Cell Signaling), total-MEK1/2 (1 : 1000; Cell Signaling), p-Akt (1 : 1000; Cell Signaling), total-Akt (1 : 1000; Cell Signaling), p-CAMKII (1 : 1000; #3362 Cell Signaling), total-CAMKII (1 : 1000; Cell Signaling), p-CAMKIV (1 : 1000; Cell Signaling), glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1 : 300; Millipore, Schwalbach, Germany). Secondary antibodies goat Anti-Mouse IgG-HRP (1 : 5000; Calbiochem, Darmstadt, Germany) or goat Anti-Rabbit IgG-HRP (1 : 5000; Calbiochem) were used. Blots were detected with SuperSignal West FEMTO Maximum Sensitivity Substrate (Thermo, Scientific, Dreieich, Germany) and the Molecular Imager ChemiDoc XRS System (Biorad, Munich, Germany). Densitometric measurements were obtained using Quantity One Software (Biorad).

ELISA

The Ras Activation ELISA Assay Kit was conducted according to the manual of Millipore (Millipore, Darmstadt, Germany). Briefly, PC12 cells were plated at a density of 1 × 107 cells/mL and were stimulated with hyperforin (10 μM) for 20 min. For sample preparation, cells were rinsed two times with PBS and solubilized in lysis buffer (1 mM EDTA, 0.5% Triton x-100, 5 mM NaF, 6 M urea, 10 μg/mL Leupeptin, 10 μg/mL, Pepstatin, 100 μM phenylmethylsulfonyl fluoride, 3 μg/mL Aprotinin, 2.5 mM Na4P2O7, 1 mM Na3VO4 in PBS, pH 7.2–7.4). Lysates were vortexed and stored at < −20°C until usage. Each probe was diluted 1 : 10 in assay buffer and measured as a triplicate. Afterwards protein content of the respective probe was determined and the obtained values were normalized on 1 mg/mL protein.

The p-CREB ELISA was performed according to the manual of R&D Systems GmbH (Wiesbaden-Nordenstadt, Germany; DuoSet® IC; DYC2510-2). PC12 cells were treated as described above. Each probe was diluted 1 : 10 in assay buffer and measured as a triplicate. Afterwards protein content of the respective probe was determined and the obtained values were normalized on 1 mg protein/mL.

Statistical analyses

Data are means of six independent experiments. Data are given as mean ± SEM. For statistical comparison using PRISM4 (GraphPad Software, La Jolla, CA, USA), Students t-test was used. p values less than 0.05 were considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References

Hyperforin activates the Ras/ERK/PI3K cascade in PC12 cells

In PC12 cells and CA1 neurons, we previously showed that hyperforin mediates neuritic outgrowth and spine morphology changes via TRPC6 channels (Leuner et al. 2007, 2013). However, the signaling cascade involved in hyperforin mediated neuritic differentiation is not finally understood. Therefore, we first investigated different calcium regulated key proteins which are involved in neuritic outgrowth (Agell et al. 2002). The Ras/MAPK pathway and Akt/PI3K cascade were tested. We focused on MEK and ERK as two important kinases of the MAPK pathway involved in PC12 cell differentiation (Takeda and Ichijo 2002). Furthermore, PI3K and Akt are also known effectors of Ras and are important regulators of PC12 cell differentiation (Chung et al. 2010).

Two different approaches were used. First, PC12 cells were stimulated with hyperforin to investigate Ras activity, phosphorylation of MEK, ERK, and Akt. Ras activity was measured using a specific ELISA detecting RAS in its activated GTP-bound state binding to its downstream kinase RAF-1. Hyperforin induced a significant activation of Ras in PC12 cells (Ras activation in %, ctl 100.1 ± 1.01, n = 6 vs. + hyperforin 133.7 ± 14.07, n = 6, p = 0.038; Fig. 1a). Furthermore, hyperforin mediated a significant phosphorylation of MEK [pMEK/total-MEK (% of ctl); ctl 94.8 ± 11.26, n = 6 vs. + hyperforin 151.8 ± 14.95, n = 6, p = 0.012; Fig. 1b], ERK [pERK/total-ERK (% of ctl) ctl 99.8 ± 11.41, n = 6; vs. + hyperforin 227.1 ± 15.66, n = 6, p < 0.001; Fig. 1c], and Akt [pAkt/total-Akt (% of ctl), ctl 100.1 ± 0.04, n = 6 vs. + hyperforin 241.8 ± 60.21, n = 6, p = 0.0432; Fig. 1d] detected by western blotting.

image

Figure 1. Hyperforin mediates neurite outgrowth via Ras, MEK/ERK, and PI3K/Akt cascade. (a) PC12 cells were incubated with hyperforin (10 μM) for 20 min. Afterwards, cells were harvested and cell lysates were used to determine active Ras. Error bars indicate means ± SEM (n = 6; unpaired t-test, *p < 0.05). (b, c, d) PC12 cells were stimulated for 20 min with hyperforin (10 μM), lysed and analyzed by western blotting using specific antibodies against p-MEK1/2 (b), p-ERK1/2 (c), p-Akt (d). Comparability was achieved by normalizing on total-MEK1/2, total-ERK1/2, and total-Akt levels. Shown is a representative blot from a single experiment that was repeated six times. Error bars indicate means ± SEM (n = 6; unpaired t-test, ***p < 0.001, *p < 0.05). (e) For neurite outgrowth assays, PC12 cells were incubated for 3 days concomitantly with hyperforin (0.3 μM) in the absence or prescence of the specific MAPK-inhibitors PD98059 (10 μM), U0126 (10 μM), the PI3K inhibitor LY294002 (10 μM), or the Akt inhibitor Akt1/2 (1 μM). Illustrated are the neurite lengths in μm (Ø 10 neurites) in comparison to hyperforin (0.3 μM) stimulated cells or to untreated control cells. Error bars indicate means ± SEM (n = 5; unpaired t-test, ***p < 0.001). (f) Shown are representative images of PC12 cells with neurite extensions.

Download figure to PowerPoint

Second, PC12 cells were incubated for 3 days with hyperforin to induce optimal neurite outgrowth in PC12 cells (Leuner et al. 2007, 2010) in the presence or absence of the specific MEK inhibitors PD98059 (Pang et al. 1995) and U0126 (Favata et al. 1998), the PI3K inhibitor LY294002 (Brunn et al. 1996) and the Akt inhibitor Akt1/2 (Fig. 1e and f). Hyperforin mediated neuritic outgrowth (ctl 28.03 ± 1.43, n = 6 vs. + hyperforin 42.69 ± 1.07 μm, n = 6, p < 0.001) was inhibited to control levels by the two specific MEK inhibitors PD98059 (28.99 ± 0.81, n = 6), and U0126 (22.42 ± 1.26, n = 6). Furthermore, the PI3K inhibitor LY294002 (29.82 ± 1.32, n = 6) and Akt inhibitor Akt1/2 (29.82 ± 1.32, n = 6) also lead to a strong reduction of neuritic length when incubated concomitantly with hyperforin. To test whether the inhibitors by themselves also inhibit neurite outgrowth in control cells, we selected one inhibitor of each signaling cascade, PD 98059 and LY294002. Both inhibitors showed a small but significant effect on neurite outgrowth if not combined with hyperforin (data not shown). This effect was expected because even control cells treated with only 2% serum show a slight differentiation which is induced by the growth factors in the culture medium which activate similar differentiation pathways such as NGF.

Hyperforin induced CREB phosphorylation is dependent on MEK/ERK and PI3K/Akt

To further characterize the hyperforin activated signaling cascade, we investigated the downstream targets of MEK/ERK and PI3K/Akt, the phosphorylation of the transcription factor CREB at Ser133. Hyperforin induced a pronounced CREB phosphorylation at Ser133 which was determined using CREB ELISA (p-CREB as % of ctl; ctl 105.8 ± 4.7, n = 6 vs. + NGF 285.9 ± 33.8, n = 6, p < 0.001, or vs. + hyperforin 190.7 ± 34.13, n = 6, p = 0.033; Fig. 2a) and a western blot analysis (p-CREB/total-CREB in % of ctl, ctl 102.61 ± 6.9, n = 6 vs. + NGF 317.34 ± 29.41, n = 6, p < 0.001, or vs. + hyperforin 230.55 ± 29.81, n = 6, p = 0.002; Fig. 2b). Compared to NGF, the effect was slightly less pronounced. To test whether CREB phosphorylation is depending on MEK/ERK and PI3K/Akt, the respective inhibitors were pre-incubated for 10 min and hyperforin (10 μM) was added (pCREB/total-CREB normalized on the hyperforin induced CREB phosphorylation, + hyoerfprin 100.12 ± 4.92, n = 6; Fig. 2c). The MEK inhibitors PD98059 (80.75 ± 3.95, n = 6, p = 0.012) and U0126 (78.01 ± 5.22, n = 6, p = 0.015) significantly inhibited CREB phosphorylation. Furthermore, LY294002 (82.25 ± 4.59, n = 6, p = 0.024) and Akt1/2 (76.5 ± 6.81, n = 6; p = 0.019) also reduced hyperforin mediated CREB phosphorylation. However, the effects of the inhibitors on hyperforin mediated phosphorylation of the respective proteins are rather modest and range between 20 and 25%. These findings might be because of the fact that both pathways are involved in hyperforin mediated phosphorylation of CREB and that there might be an additional pathway involved such as CAMK pathway.

image

Figure 2. Hyperforin induces via the MEK/ERK and PI3K/Akt pathways cAMP-response element binding protein (CREB) phosphorylation. For p-CREB ELISA (a) and western blot analysis (b), PC12 cells were stimulated for 20 min with hyperforin (10 μM) or nerve growth factor (NGF) (50 ng/mL). PC12 cells were lysed and analyzed by ELISA or western blotting using p-CREB antibodies. Comparability was achieved by normalizing on total-CREB-levels. Shown is a representative blot from a single experiment that was repeated six times. Error bars indicate means ± SEM (n = 6; unpaired t-test, ***p < 0.001, **p < 0.01). (c) Hyperforin mediated CREB phosphorylation is depending on MEK/ERK or PI3K/Akt. PC12 cells were pre-incubated for 15 min with the MEK inhibitors PD98059 (10 μM) and U0126 (10 μM), the PI3K inhibitor LY294002 (10 μM), or the Akt inhibitor Akt1/2 (1 μM). Afterwards, hyperforin (10 μM) was added. Data are normalized on total-CREB levels and are exhibited as the percentage of the hyperforin mediated CREB phosphorylation. Error bars indicate means ± SEM (n = 6; unpaired t-test, **p < 0.01, *p < 0.05).

Download figure to PowerPoint

TRPC6 channels are required for MEK/ERK/Akt/CREB phosphorylation

To test if functional TRPC6 channels are required for hyperforin-induced phosphorylation of MEK, ERK, Akt, and CREB we over-expressed pore-dead dominant-negative mutant (dnTRPC6-YFP) in PC12 cells resulting in an inactivation of TRPC6 in transfected cells (Leuner et al. 2007). In dnTRPC6-YFP expressing PC12 cells the hyperforin induced phosphorylation of MEK (pMEK/total-MEK in %, ctl + hyperforin 105.02 ± 7.14, n = 6 vs. dnTRPC6YFP + hyperforin 63.00 ± 8.47, n = 6, p = 0.0035; Fig. 3a), ERK (pERK/total-ERK in %, ctl + hyperforin 109.02 ± 10.17, n = 6 vs. dnTRPC6YFP + hyperforin 65.60 ± 9.87, n = 6, p = 0.012; Fig. 3b), Akt (pAkt/total-Akt in %, ctl + hyperforin 106.42 ± 6.54, n = 6 vs. dnTRPC6YFP + hyperforin 71.23 ± 10.08, n = 6, p = 0.019; Fig. 3c), and CREB (pCREB/total-CREB in %, ctl + hyperforin 94.08 ± 5.73, n = 6 vs. dnTRPC6YFP + hyperforin 47.13 ± 12.44, n = 6, p = 0.0064; Fig. 3d) is significantly reduced.

image

Figure 3. Hyperforin mediated phosphorylation of MEK/ERK, Akt, and cAMP-response element binding protein (CREB) is dependent on transient receptor canonical (TRPC)6 channels. PC12 cells were transfected for 48 h with dnTRPC6-YFP plasmid and stimulated for 20 min with hyperforin (10 μM). Western blots using p-MEK1/2, p-ERK1/2, p-Akt, or p-CREB antibody were compared with hyperforin (10 μM) stimulated untransfected PC12 cells. Comparability was achieved by normalizing on total-MEK1/2, total-ERK1/2, total-Akt, or total-CREB protein load respectively. Error bars indicate means ± SEM (n = 6; unpaired t-test, *p < 0.05, **p < 0.01).

Download figure to PowerPoint

Hyperforin also leads to MEK/ERK/Akt/CREB phosphorylation in primary hippocampal neurons

Hyperforin not only induces TRPC6 mediated calcium influx and neurotrophic effects in PC12 cells but also CA1 neurons in organotypic cultures of post-natal hippocampal slices (Leuner et al. 2013). Especially the neurotrophic effect of hyperforin on hippocampal neurons is important considering the exceptional role of the hippocampus in the pathophysiology of depression (Pittenger and Duman 2008). Therefore, we first confirmed that hyperforin also led to synaptic plasticity changes in primary hippocampal neurons by altering the number of synapses. Synapsin, a pre-synaptic marker, and PSD-95, a post-synaptic marker, were used to investigate the effects of TRPC6 activation by hyperforin. Using confocal microscopy, we could show that hyperforin significantly increased the number of synapsin (synapsin positive pre-synapses/10 μm dendrite, ctl 2.25 ± 0.07, n = 6 vs. + BDNF 2.91 ± 0.08, n = 6, p < 0.001 or vs. + hyperforin 2.62 ± 0.06, n = 6, p = 0.0035) and PSD-95 positive bona fide synapses (PSD95 positive pre-synapses/10 μm dendrite, ctl 0.78 ± 0.03, n = 6, vs. + BDNF 1.15 ± 0.03, n = 6, p < 0.001, or vs. + hyperforin 0.95 ± 0.03, n = 6, p = 0.0018) in mature primary hippocampal neurons (Fig. 4a–d). The hyperforin mediated effect is comparable to the increase in bona fide synapses by BDNF. In addition, TRPC6 activation led to a similar activation profile of signal transduction pathways as assessed phosphorylation levels of MEK, ERK, Akt, and CREB (Fig. 4e–h).

image

Figure 4. MEK/ERK, PI3K, and cAMP-response element binding protein (CREB) are also downstream targets of transient receptor canonical (TRPC)6 channels in primary hippocampal neurons. Synapse density as number of synapsin positive pre-synapses (a) or PSD-95 positive post-synapses (c) per 10 μM of dendrite in mature hippocampal neurons representative example of synapsin (b) or PSD-95 (d) stained bona fide synapses of mature hippocampal neurons and. Primary hippocampal neurons were stimulated for 20 min with hyperforin (10 μM), lysed and analyzed by western blotting using specific antibodies against p-MEK1/2 (e), p-ERK1/2 (f), p-Akt (g), and p-CREB (h). Comparability was achieved by normalizing on total-MEK1/2, total-ERK1/2, total-Akt, or total-CREB protein levels respectively. Shown is a representative blot from a single experiment that was repeated six times. Error bars indicate means ± SEM (n = 6; unpaired t-test, ***p < 0.001, **p < 0.01, *p < 0.05).

Download figure to PowerPoint

TRPC6 activation by hyperforin also activates CAMKIV in PC12 cells and primary hippocampal neurons

In addition to MAPK and PI3K signaling cascades, CAMK is an important target of calcium finally resulting in CREB phosphorylation. In PC12 cells, the role of CAMKII and CAMKIV for Ca2+ mediated neuritic outgrowth was not finally clarified (Miranti et al. 1995; Beitner-Johnson et al. 1998). Therefore, western blot analysis was conducted with specific antibody for p-CAMKIV, CAMKII, and p-CAMKII. TRPC6 activation by hyperforin leads to an increase in p-CAMKIV levels in PC12 cells (pCAMKIV/GAPDH in % of ctl; ctl 93.16 ± 10.17, n = 6, vs. + hyperforin 163.6 ± 21.41, n = 6, p = 0.014; Fig. 5a). For CAMKII protein, we found a strong protein signal in PC12 cells. However, after treatment with hyperforin no specific CAMKII phosphorylation could be detected (data not shown). The described effect on CAMKIV is dependent on TRPC6 because over-expression of TRPC6dn led to a significant reduction of p-CAMKIV levels in PC12 cells (pCAMKIV/GAPDH in % of ctl, ctl 112.64 ± 13.7, n = 6, vs. + hyperforin 70.22 ± 5.56, n = 6, p = 0.088) Fig. 5b). Furthermore, we obtain similar results in primary hippocampal neurons (pCAMKIV/GAPDH in % of ctl, ctl 96.33 ± 3.84, n = 6 vs. + hyperforin 132.73 ± 12.03, n = 6, p = 0.016; Fig. 5c) regarding CAMKIV phosphorylation.

image

Figure 5. Hyperforin mediated transient receptor canonical (TRPC)6 activation induces CAMKIV phosphorylation in PC12 cells and primary hippocampal neurons. (a) PC12 cells were treated for 20 min with hyperforin (10 μM), and western blot analysis with a specific p-CAMKIV antibody was performed. Comparability was achieved by normalizing on GAPDH protein levels. (b) PC12 cells were transfected for 48 h with TRPC6-DN-YFP plasmid and were then stimulated for 20 min with hyperforin (10 μM). (c) Primary hippocampal neurons were treated for 20 min with hyperforin (10 μM) and p-CAMKIV was detected using western blotting. Error bars indicate means ± SEM (n = 6; unpaired t-test, *p < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References

In this study, we describe novel findings into the signaling cascade of TRPC6 mediated neuritic outgrowth in PC12 cellls and primary hippocampal neurons. The TRPC6 activator hyperforin leads to the activation of three signal cascades downstream of calcium cumulating in the phosphorylation of CREB: Ras/MEK/ERK, PI3K/Akt, and CAMKIV. The activation of these pathways seems to be essential for TRPC6 mediated changes in synaptic plasticity in PC12 cells and primary hippocampal neurons.

TRPC6 channels are highly expressed in the CNS (Riccio et al. 2002). They are involved in cell survival and protection against apoptosis (Du et al. 2010; Lin et al. 2013), dendritic and neurite outgrowth (Leuner et al. 2007; Tai et al. 2008), spine morphology changes (Leuner et al. 2013), and the formation of excitatory synapses (Zhou et al. 2008). However, the TRPC6 activated signaling cascades which translate calcium transients into structural changes are not finally clarified yet. Therefore, we decided to use the TRPC6 activator hyperforin which was shown to mediate it's neurotrophic effects via TRPC6 channels in PC12 cells and primary hippocampal neurons (Leuner et al. 2007, 2013).

We show that hyperforin induced TRPC6 activation leads to Ras activation. Our data are supported by findings in cerebellar granule neurons where TRPC6 over-expression also leads to Ras activation (Jia et al. 2007). Ras plays an essential role in the regulation of proliferation and differentiation in the CNS (Cullen and Lockyer 2002; Cullen 2006). Ras activity is not only regulated by growth factor receptors but also by elevation of intracellular calcium (Cullen and Lockyer 2002; Cullen 2006). Interestingly, transient calcium influx via L-type voltage dependent calcium channels or NMDA receptors leads to Ras dependent ERK activation finally regulating synaptic plasticity in form of long term potentiation (Dolmetsch et al. 2001; Jin and Feig 2010).

Downstream of Ras, hyperforin induced TRPC6 activation leads to the activation of both the MEK/ERK and the PI3K/Akt pathways. Hyperforin mediated neuritic outgrowth is inhibited by the MEK/ERK inhibitors PD98059 and U0126 as well as the PI3K/Akt inhibitors LY294002 and Akt1/2. MEK and PI3K are both downstream targets of Ras (Agell et al. 2002) and are key regulators of neuritic outgrowth in PC12 cells (Riese et al. 2004), spine morphology changes and dendritic outgrowth in hippocampal neurons (Alonso et al. 2004; Kumar et al. 2005) and also activated by classical anti-depressants (Cowen 2007). The MAPK pathway seems to be involved in the pathophysiology of depression (Duman et al. 2007). Agents which inhibit the MAPK pathway show negative effects in behavioral models for depression such as the forced swim test or the learned helplessness paradigm (Duman et al. 2007). In post-mortem tissue from depressed suicide subjects, decreased ERK activity, and ERK expression was detected. Recently, a whole genome expression profiling of post-mortem tissue from depressed patients revealed the enhanced expression of a negative regulator of the MAPK cascade, the mitogen-activated protein kinase phosphatase-1 (MKP-1) (Duric et al. 2010). Importantly, anti-depressant treatment with fluoxetine normalizes MKP-1 expression and improves depressive behavior.

Activation of PI3K/Akt and ERK/MEK via TRPC6 channels finally results in CREB phosphorylation. The inhibitors of PI3K/Akt and ERK/MEK significantly inhibited hyperforin mediated CREB phosphorylation. The role of CREB in TRPC6 mediated dendritic outgrowth and spine density is supported by findings that a dominant negative mutant of CREB completely inhibited the TRPC6 mediated dendritic growth in hippocampal neurons and increased spine density in CA1 neurons (Tai et al. 2008; Zhou et al. 2008). In addition, CREB phosphorylation induced by hyperforin was also shown by other groups in different tissues (Gibon et al. 2013; Lin et al. 2013). Chemical anti-depressants such as serotonin uptake inhibitors or tricyclic drugs also enhance CREB phosphorylation, which is considered to be strongly involved in anti-depressant response (Pittenger and Duman 2008). Viral over-expression of CREB in the hippocampus shows anti-depressant effects in behavioral models of depression such as the forced swim test or the learned helplessness paradigm (Chen et al. 2001). Furthermore, clinical improvement of patients is strongly associated with the increase in p-CREB when treated with anti-depressants (Koch et al. 2005, 2009). However, there might be an additional role of BDNF in the neurotrophic effect of hyperforin in primary hippocampal neurons. The hyperforin mediated intracellular Ca2+ elevations could also cause BDNF release from neurons, which in turn would activate all the signaling pathways studied here (Li et al. 2010). This hypothesis will be investigated in future experiments.

Finally, we investigated if the hyperforin induced signaling cascade is specifically mediated by TRPC6. TRPC6 was inactivated using a pore dead mutant as described before (Leuner et al. 2007). Hyperforin mediated phosphorylation of MEK/ERK, PI3K/Akt, and CREB was significantly inhibited. These findings are in line with our previous results showing that calcium influx via TRPC6 channels is essential for hyperforin mediated neuritic outgrowth. Our findings are supported by the group of Wang which demonstrated that TRPC6 channels mediated formation of excitatory synapses and dendritic growth is induced by a CREB (Tai et al. 2008; Zhou et al. 2008). Knockdown of TRPC6 channels resulted in a pronounced loss of excitatory synapse and spine density. An over-expression of TRPC6 induced a significant increase in dendritic spine density in hippocampal CA1 neurons. In cerbellar granule cells, BDNF mediated cell protection against serum deprivation is mediated by calcium influx via TRPC3 and TRPC6 channels and CREB phosphorylation through the Ras/MEK/ERK pathway (Jia et al. 2007). This mechanism is comparable with neuronal survival as a consequence of L-type VGCC in response to high K+-induced depolarization. In contrast to our findings, Akt was not a downstream target of TRPC3 and TRPC6 activation by BDNF. This discrepancy might be explained by the different stimuli BDNF and hyperforin. BDNF activates TRPC3 and TRPC6 channels via the activation of PLC and the generation of DAG. Hyperforin might directly activate TRPC6 channels and therefore show another profile of signal transduction (please also see Fig. 6) Furthermore, TRPC6 degradation and subsequently decreased CREB phosphorylation may contribute to ischemic brain damage (Du et al. 2010; Lin et al. 2013). Prevention of cleavage of TRPC6 in a rat model reduced infarct size and improved behavioral performance of the rats.

image

Figure 6. Comparison between brain derived neurotrophic factor (BDNF) and hyperforin mediated effects on transient receptor canonical (TRPC)6 activated signaling cascade.

Download figure to PowerPoint

Finally, we focused on CAMK II and IV which are also important targets for Ca2+ mediated neurite outgrowth resulting in CREB phosphorylation (Fig. 5). Hyperforin mediated TRPC6 activation leads to CAMKIV phosphorylation in PC12 cells and primary hippocampal neurons. Importantly, TRPC6 channel over-expression in hippocampal neurons also induced CAMKIV phosphorylation and promotes via a CAMKIV/CREB pathway formation of spines and excitatory synapses (Tai et al. 2008; Zhou et al. 2008). Interestingly, we were not able to detect CAMKII phosphorylation by hyperforin in PC12 cells even if we had a robust signal for CAMKII protein signal supporting previous data from other groups (Miranti et al. 1995; Beitner-Johnson et al. 1998). In the brain, CAMKIV is expressed in regions such as the hippocampus, and the amygdala and is involved in fear memory, synaptic plasticity and age associated memory deficits (Ho et al. 2000; Wei et al. 2002; Fukushima et al. 2008; Takao et al. 2010). In CAMKIV−/− mice, decreased long-term potentiation in CA1 neurons of the hippocampus was detected (Ho et al. 2000). In aged mice, decreased CAMKIV levels were associated with memory decline which was rescued by the over-expression of CAMKIV (Fukushima et al. 2008). Therefore, phosphorylation of CAMKIV might not only be important for the anti-depressant effects of St. John's wort but also for memory enhancing properties of the extract in animal models (Dinamarca et al. 2006; Trofimiuk et al. 2010).

Taken together, we show that hyperforin induced TRPC6 channel activation mediates neuritic outgrowth via several mechanisms comprising Ras, ERK, PI3K, CAMKIV activation finally resulting in CREB phosphorylation in PC12 cells and primary hippocampal neurons (Fig. 6). This described signaling cascade for hyperforin might be of special relevance for the anti-depressant effects of St. John's wort.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References

We thank Dr. Willmar Schwabe for providing hyperforin, standardized SJW extract, and the different constituents of SJW extract. In addition, we thank the Emerging Fields Initiative, Friedrich-Alexander-Univeristät Erlangen, for financial support.

J.H.H. contributed to the conception and design, acquisition, analysis and interpretation of data. A.M.S. contributed to the acquisition, analysis, and interpretation of data. W.E.M. revised critically the article. K.L. contributed to the conception and design as well as the interpretation of data. She drafted the manuscript and gave the final approval of the version to be published.

Conflict of interest

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References

W.E.M. is a paid consultant of several pharmaceutical companies, including Dr. Willmar Schwabe (Germany), but receives no royalty (cash or otherwise) from sales of any products.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Conflict of interest
  8. References