MKP3 eliminates depolarization-dependent neurotransmitter release through downregulation of L-type calcium channel Cav1.2 expression

Cell Calcium 53 (2013) 224–230

Contents lists available at SciVerse ScienceDirect

Cell Calcium journal homepage: www.elsevier.com/locate/ceca

MKP3 eliminates depolarization-dependent neurotransmitter release through downregulation of L-type calcium channel Cav1.2 expression Ole V. Mortensen ∗ Department of Pharmacology and Physiology, Drexel University College of Medicine, Philadelphia, PA 19102, USA

a r t i c l e

i n f o

Article history: Received 3 August 2012 Received in revised form 6 December 2012 Accepted 24 December 2012 Available online 18 January 2013 Keywords: Voltage-gated calcium channel Gene expression Map kinase Signaling Phosphatase Exocytosis pc12 cells

a b s t r a c t Release of neurotransmitters is a fundamental and regulated process that is essential for normal brain functioning. Regulation of this process is potentially important for any neuronal process, and disruption of the release process may contribute to the pathophysiology associated with psychiatric diseases. In this work it is shown that expression of the negative regulator of mitogen-activated protein kinase (MAPK) signaling the MAPK phosphatase MKP3/DUSP6 eliminates depolarization-dependent release of dopamine in rat PC12 cells. Pharmacologic interventions with latrotroxin (LTX) or A23187, which make the cells permeable to calcium, reestablish the dopamine release. Calcium imaging also reveals that calcium influx is impaired in MKP3-expressing cells. Because acute pharmacologic inhibition of MAPKs has no effect on dopamine release in naïve PC12 cells, the MKP3-mediated elimination of neurotransmitter release must be caused by a long-term process, such as changes in gene expression. In support of this the expression of the L-type calcium channel cav1.2 alpha subunit (Cacna1c) is decreased in MKP3-expressing PC12 cells. With the reintroduction of cav1.2 expression, neurotransmitter release is restored in the MKP3-expressing PC12 cells. Thus, MKP3 expression reduces neurotransmitter release by decreasing the expression of cav1.2. Because MKP3 is increased when neuronal activity is elevated, this process could play a role in regulating neurotransmitter homeostasis. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Vesicular neurotransmitter release is a key event in neuronal intercellular communication and has been investigated extensively [1]. Neurotransmitter release from intracellular vesicular stores is a fast, highly regulated process. Because of its fundamental importance for multiple biological processes, understanding the detailed mechanisms of this process is of great biological and medical interest. Furthermore, subtle changes in the regulation of neurotransmitter release could have an impact on information coding in neural macro- and microcircuits and could contribute to the pathophysiology associated with psychiatric diseases [2]. The regulated release of neurotransmitters is tightly dependent on calcium entering the cells following the arrival of a depolarizing action potential and the activation of voltage-dependent calcium channels. The rise in intracellular calcium triggers the release of neurotransmitters from vesicles. The core protein release machinery is well studied and includes the voltage-dependent calcium channels and SNARE proteins on the plasma and vesicular

∗ Tel.: +1 215 762 6947; fax: +1 215 762 2299. E-mail addresses: [email protected], [email protected] 0143-4160/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ceca.2012.12.004

membranes [3]. Calcium is the main signal triggering the release, but several studies have established that intracellular signaling pathways, among others the phosphorylation of several of the abovementioned presynaptic proteins [4–6], are involved in regulating and fine tuning the release process. One kinase family among these signaling pathways that has been implicated in neurotransmitter release is the MAPKs ERK1/2 [7]. Most studies have found that long-term activation of ERK1/2 with growth factors such as brain-derived neurotrophic factor (BDNF) results in enhanced release [8,9]. Conversely, short-term inhibition of ERK1/2 also produces enhanced neurotransmitter release as a result of increased calcium flux through L-type calcium channels [10]. The endogenous negative regulators of MAPK signaling are the MAPK phosphatases DUSP/MKP (DUSP for dual specificity phosphatase and MKP for MAPK phosphatase), which inactivate the kinase through the dephosphorylation of a threonine and a tyrosine residue [11]. Some MKPs are found to be upregulated by neuronal activity [12,13] and could therefore be involved in homeostatic plasticity within the presynaptic neuron. In this study I have tested and investigated the role of MKPs on neurotransmitter release. I demonstrate that increasing MAPK phosphatase 3 (MKP3) expression will dramatically downregulate neurotransmitter release from large dense core vesicles in rat pheochromocytoma PC12 cells. Importantly, I establish that MKP3 attenuates exocytosis

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

via a downregulation of the expression of the voltage-dependent calcium channel Cav1.2 and hypothesize that this process could indeed provide a physiological mechanism for regulating neuronal activity. 2. Materials and methods 2.1. Materials [3 H]-dopamine and [3 H]-serotonin were obtained from Perkin Elmer (Boston, MA, USA). Reagents for buffers, inhibitors, and other chemicals were purchased from Sigma-RBI (St. Louis, MO, USA). Cell culture media, fetal bovine serum (FBS), penicillin/streptomycin, and Dulbecco phosphate-buffered saline (D-PBS) were purchased from Life Technologies (Carlsbad, CA, USA). PC12 cells were maintained in Dulbecco modified Eagle medium (3.5 g/L glucose) supplemented with FBS (10% each) at 37 ◦ C with 5% CO2 . For expression and production of stable PC12 cell lines, MKP1 or MKP3 was subcloned into pIRES2 to enable color selection, and stably expressing colonies were isolated in the presence of G418 (1 mg/mL). For co-transfection experiments, cells were harvested the day before transfection and plated in 24-well plates. After 24 h, cells were transiently transfected with combinations of human serotonin transporter cDNA and various calcium channel cDNAs using Lipofectamine (Life Technologies) according to the manufacturer’s protocol. The cDNAs for Cav1.2 [14] (Addgene plasmid 26572), and Cav2.2 [15] (Addgene plasmid 26568) were kind gifts from Dr. Diane Lipscombe (Brown University). Experiments were performed 2 days after transfection. 2.2. Neurotransmitter release from PC12 cells On the day of the experiment, the culture medium was replaced by 250 ␮L of complete medium containing 50 nM tritiated serotonin or dopamine (Perkin Elmer) to preload the cells. After 1 h at 37 ◦ C, the cells were washed with 0.5 mL PBS. Release experiments were performed at 25 ◦ C, in 0.25 mL Krebs Ringer solution (130 mM NaCl, 1.3 mM KCl, 2.2 mM CaCl2 , 1.2 mM MgSO4 , 1.2 mM KH2 PO4 , 10 mM HEPES, 10 mM d-glucose). When potassium was used to stimulate release, the solution contained 50 mM NaCl and 80 mM KCl. Following 10 min of incubation, the medium was collected and the cells were lysed and the radioactivity of medium and cells was determined by scintillation counting. 2.3. Immunoblotting to detect phosphorylation state of ERK For immunoblotting experiments PC12 cells were lysed in 2× sample buffer. All samples were separated by SDS-PAGE and immunoblotted. Antibodies from Cell Signaling Technology (Danvers, MA) were used to detect the level of and the activation state of ERK1/2. An antibody against beta-actin was used as loading control. 2.4. Calcium imaging The calcium indicator Rhod-3 AM (Life Technologies) was used to image calcium flux in naïve and MKP3-expressing PC12 cells. Cells were loaded with 10 ␮M Rhod-3 AM according to the manufacturer’s protocol for 60 min in the dark, then washed and incubated in incubation buffer containing 2.5 mM probenecid for 60 min in the dark. Fluorescence emission was detected at 580 nm. To trigger calcium flux, the same buffer compositions were used as were used in the release experiments. The flow rate was 1 mL/min.

225

2.5. Gene expression studies RNA was isolated from PC12 cell lines according to the manufacturer’s protocol using RNeasy columns (Qiagen, Valencia, CA). cDNAs were produced using the High Capacity RNA-to-cDNA Kit (Life Technologies). Taqman-based gene expression assays were performed on an Applied Biosystem (Life Technologies) 7900HT Fast Real-Time PCR System using TaqMan Gene Expression Assay against glyceraldehyde-3-phosphate dehydrogenase (GAPDH), Cav1.2 (CACNA1C), and Cav2.2 (CACNA1B) (Life Technologies). Relative expression levels were calculated using the standard formula, 2−CT subtracting threshold cycles CT of GAPDH from threshold cycles CT of Cav1.2 and Cav1.3 respectively. 3. Results 3.1. MAPK phosphatase 3 (MKP3) expression reduces depolarization-dependent neurotransmitter release To investigate whether MAPK signaling is involved in vesicular release of neurotransmitter from large dense core vesicles in PC12 cells, stable cell lines were established that express MKP1 and MKP3, respectively. Three different secretagogues were used to study the effect of MKP expression on neurotransmitter release. These include the calcium ionophore A23187 (1 ␮M), the black widow spider toxin LTX (1 nM), which acts as a calcium ionophore but which also has other protein interaction-mediated effects on release [16], and finally depolarized PC12 cells, incubated in a buffer containing 80 mM potassium to activate voltage-dependent calcium channels. Cells were loaded with tritiated dopamine and the effect of the above three secretagogues on released dopamine was measured. The results from these experiments (Fig. 1) revealed a significant involvement of MKP3 in regulated neurotransmitter release. As expected, all three manipulations resulted in a significant release of dopamine in naïve and MKP1-expressing PC12 cells (Fig. 1). In MKP3-expressing cells, the two calcium ionophores A23187 and LTX, comparable to the two other cell types, triggered release of dopamine. Importantly and differently from naïve and MKP1-expressing PC12 cells and revealing a specific role for MKP3 in depolarization dependent neurotransmitter release, high potassium-mediated depolarization was unable to trigger any vesicular release of dopamine in the MKP3-expressing PC12 cells. 3.2. Neurotransmitter release is vesicular and not transporter mediated Depolarization dependent release of neurotransmitter suggested that vesicular release processes were compromised in the MKP3 expressing PC12 cells, but PC12 cells also endogenously express the norepinephrine transporter (NET) that in some situations will work in reverse and release intracellular neurotransmitters such as dopamine [17]. To examine whether the released neurotransmitter observed in Fig. 1 is a result of reverse NET transporter function, NET was inhibited with the potent classical tricyclic antidepressant desipramine (Fig. 2A). The Ki of desipramine for inhibiting NET is in the low nanomolar range [18]. At concentrations of 100 ␮M there was no effect on the dopamine release triggered by high potassium. Conversely, the calcium chelator EGTA at a 5mM concentration completely blocked the release of dopamine by high potassium (Fig. 2A). In addition to showing complete calcium dependence of the release process, the ladder result further eliminates the involvement of NET as NET function is not calcium dependent in PC12 cells [19]. In conclusion, the released dopamine is a result of only calcium-dependent release of neurotransmitter from the large dense core vesicles in the PC12 cells.

226

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

Fig. 1. Depolarization-dependent vesicular release is absent in MKP3-expressing PC12 cells. Naïve PC12 cells (PC12) or PC12 cells stably expressing MKP1 (MKP1-PC12) or MKP3 (MKP3-PC12) were loaded with radiolabeled dopamine and exocytosis was initiated by adding Krebs Ringer buffer containing vehicle, the calcium ionophore A23187 (1 ␮M), latrotoxin (LTX) (1 nM), or 80 Mm K+ . All experiments were performed at least three times in triplicate. The amount of dopamine released was significantly different from vehicle (p < 0.05) using two way ANOVA with Bonferroni post hoc test comparing with naïve cells except for 80 mM K+ in MKP3-expressing cells.

3.3. Inhibition of MAPK signaling does not affect neurotransmitter release The main physiological function of MKP3 is to specifically inhibit the ERK1/2 MAPKs by dephosphorylation. It would therefore be expected that pharmacological inhibition of ERK1/2 kinase activity would have similar effects as overexpression of MKP3. To test this premise specific MEK1/2 inhibitors U0126 and SL327 were employed. MEK1/2 is directly upstream of ERK1/2 in this MAPK signaling pathway. Surprisingly, inhibiting MEK1/2, using U0126 at 10 ␮M for 30 min prior to the release experiments (Fig. 2B), had no effect on high potassium-mediated dopamine release. Treatment of PC12 cells with U0126 at this concentration completely eliminated the base level phosphorylation of ERK1/2 observed by immunoblotting (Fig. 2C). Another MEK1/2 inhibitor SL327 was also tested but with a different result. At a concentration (10 ␮M) where ERK1/2 was inhibited a small but significant inhibition of neurotransmitter release was observed (Fig. 2B and C). Because U0126 completely blocks ERK1/2 activation and does not affect

neurotransmitter release, U0126 and SL327 must work through different mechanisms. As all activated ERK1/2 that is detectable with this assay is inhibited by U0126, it could be speculated that SL327 inhibits neurotransmitter release through an ERK1/2 independent mechanism. However, it cannot be completely ruled out that SL327 inhibits a subpopulation of MEK1/2 kinases that are not detectable with this assay and that are not affected by U0126. It is of interest to note that another MEK1/2 inhibitor, PD98059, has been shown to affect glutamate release in synaptosomes through a MEK1/2 independent direct inhibition of N- and P-/Q-type voltage-gated Ca2+ channels [20]. 3.4. Calcium influx is absent in MKP3-expressing cells Because calcium ionophores such as LTX and A23187 triggered neurotransmitter release in MKP3-expressing PC12 cells, the effects of MKP3 expression on depolarization-dependent dopamine release could be caused by changes in calcium flux. To examine this possibility, cell calcium imaging was performed. Naïve and

Fig. 2. (A) Neurotransmitter release is calcium dependent. Naïve PC12 cells were loaded with radiolabeled dopamine and exocytosis was initiated by adding Krebs Ringer buffer containing 1.3 mM K+ or 80 mM K+ . Norepinephrine transporter-dependent processes were inhibited by the inclusion of 100 ␮M desipramine. Calcium dependence was tested by chelating calcium with 5 mM EGTA. *p < 0.01 compared to vehicle using unpaired t-test. (B) Two inhibitors (U0126 and SL327) of MEK1/2 MAPK signaling do not affect neurotransmitter release in a similar manner. The MAPKs MEK1/2 were inhibited by inclusion of 10 ␮M U0126 or SL327 for 30 min prior to exocytosis assay. All experiments were performed at least three times in triplicate. ***p < 0.001 compared to vehicle using unpaired t-test. (C) A representative immunoblot showing ERK1/2 phosphorylation following 30 min preincubation with U0126 or SL327. ERK1/2 was activated with 1 ␮M phorbol 12-myristate 13-acetate (PMA).

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

227

Fig. 3. Depolarization-dependent intracellular calcium flux is absent in MKP3expressing cells. Naïve PC12 cells or PC12 cells stably expressing MKP3 (PC12-MKP3) were loaded with Rhod-3 AM and intracellular calcium was imaged following addition of 80 mM K+ containing Krebs Ringer buffer (K+ ). Flow rate was 1 mL/min. Multiple cells were assayed and pooled for analysis. All experiments were performed at least three times.

MKP3-expressing PC12 cells were loaded with the calcium indicator Rhod-3 AM and calcium flux was imaged in response to changing the perfusion buffer to high potassium. Naïve PC12 cells displayed a robust increase in intracellular calcium following perfusion of the high potassium buffer (Fig. 3). Conversely, in the MKP3-expressing PC12 cells, no increase was detectable in intracellular calcium in response to the depolarizing high-potassium perfusion buffer. The absence of increased intracellular calcium in MKP3-expressing PC12 cells would explain the decreased vesicular release as the release is triggered by increased intracellular calcium [3]. 3.5. Voltage-dependent calcium channel Cav1.2 expression is downregulated by MKP3 An explanation for the absence of a depolarization-dependent increase in intracellular calcium would be a decrease in surface expression of active voltage-dependent calcium channels. RNAs from naïve and MKP3-expressing cells were isolated and Cav1.2 and Cav2.2 mRNAs were quantified using quantitative real-time RT-PCR. In unpublished experiments using DNA microarrays to study global gene expression in naïve and MKP3-expressing cells, we have found that these two calcium channels are the only detectable calcium channel alpha subunits expressed by PC12 cells. In MKP3-expressing cells, the expression of Cav1.2 was decreased significantly, with a dramatic 90% decrease in Cav1.2 transcript (Fig. 4). Interestingly, no change in Cav2.2 expression was observed.

Fig. 4. Voltage-dependent calcium channels Cav1.2 and Cav2.2 are expressed in PC12 cells but only Cav1.2 expression is downregulated by MKP3 expression. Total RNA was isolated from naïve (PC12) and MKP3-expressing cells (PC12 + MKP3) and Cav1.2 and Cav2.2 mRNA was quantified using quantitative real-time RT-PCR. GAPDH was included as an internal control. ***p < 0.001 compared to naïve PC12 cells using unpaired t-test.

but not Cav2.2 restored the depolarization-dependent serotonin release, directly implicating the downregulation of Cav1.2 expression in the reduced dopamine release observed in MKP3-expressing cells. 3.7. MKP3 expression alters both depolarization-dependent and constitutively active calcium channel-mediated neurotransmitter release Finally, pharmacologic strategies were used to understand the contribution of calcium channels to neurotransmitter release in PC12 cells (Fig. 6). The literature is not in complete agreement regarding which calcium channels are expressed in undifferentiated PC12 cells [22–25]. Nifedipine (10 ␮M) was used to inhibit L-type Cav1.2 channels; ␻-conotoxin GVIA (1 ␮M) was used to inhibit N-type Cav2.2 channels; and cadmium (100 ␮M) was used

3.6. Ectopic expression of Cav1.2 but not Cav2.2 restores depolarization-dependent neurotransmitter release The finding of reduced Cav1.2 expression could explain the absence of increased intracellular calcium and of dopamine release in MKP3-expressing cells but does not exclude the possibility that other vesicular release processes are affected by MKP3 expression. To test whether Cav1.2 expression is the sole mediator of the reduced dopamine release in MKP3-expressing PC12 cells, voltagedependent calcium channels were reintroduced by transiently transfecting into the MKP3-expressing cells’ cDNAs for Cav1.2 and Cav2.2, respectively (Fig. 5). Because PC12 cells have low transfection efficiency, a strategy co-transfecting the individual calcium channel cDNAs with a serotonin transporter cDNA was employed [21]. Only cells that are transfected with both cDNAs were loaded if radiolabeled serotonin is used. Also, because serotonin is efficiently packaged into the vesicles in PC12 cells, release experiments can be performed using this neurotransmitter instead of dopamine. When the Cav1.2 and Cav2.2 calcium channels were reintroduced individually into MKP3-expressing PC12 cells, the expression of Cav1.2

Fig. 5. Ectopic expression of Cav1.2 but not Cav2.2 reestablishes depolarizationdependent neurotransmitter release in MKP3-expressing PC12 cells. cDNAs for Cav1.2 or Cav2.2 were co-transfected with cDNA for the serotonin transporter into MKP3-expressing PC12 cells by transient transfection. Transfected cells were loaded with radiolabeled serotonin and exocytosis was initiated by adding Krebs Ringer buffer containing vehicle or 80 mM K+ . All experiments were performed at least three times in triplicate. ***p < 0.001 compared to vehicle using unpaired t-test.

228

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

Fig. 6. MKP3 expression alters depolarization-dependent and -independent calcium channel-mediated neurotransmitter release. The contributions of L-type, N-type, and other calcium channels to the neurotransmitter release were investigated using specific (␻-conotoxin GVIA, 1 ␮M, and nifedipine, 10 ␮M) and nonspecific (cadmium, 100 ␮M) calcium channel blockers. Naïve PC12 cells (PC12) or MKP3-expressing PC12 cells transiently transfected with cav1.2 (G3 + Cav1.2) were loaded with radiolabeled dopamine and exocytosis was initiated by adding Krebs Ringer buffer containing vehicle or 80 mM K+ and calcium channel blockers. All experiments were performed at least three times in triplicate. To determine if the amount of dopamine released was significantly different from vehicle a two way ANOVA with Bonferroni post hoc test was performed (***p < 0.001).

as a nonspecific calcium channel blocker. As expected, nifedipine and cadmium blocked depolarization-mediated serotonin release in the Cav1.2-transfected MKP3-expressing cells (G3 + Cav1.2). Interestingly, cadmium blocked an apparent calcium leak in the naïve PC12 cells but not in the MKP3-expressing cells. The molecular determinant of this leak that is absent in the MKP3-expressing cells is not clear.

release triggered by depolarizing PC12 cells with high potassium. Interestingly, when release was triggered with other methods, such as treating the PC12 cells with calcium ionophores including A23187 and black widow spider LTX, the release was not affected by MKP3 expression. This specific finding that only depolarization dependent release was affected suggests that calcium flux could be compromised and this could explain the observed effect of reduced dopamine release. Fluorescent imaging was used to assess intracellular calcium levels in response to depolarization and revealed that naïve PC12 cells displayed the expected increase in intracellular calcium levels in response to depolarization, but depolarizationtriggered calcium flux was absent in MKP3-expressing PC12 cells. The main mediator of calcium flux in response to depolarization is membrane-expressed voltage-dependent calcium channels [6]. Voltage-dependent calcium channels can be divided into three main groups based on their pharmacology, including the P- and Qtype (Cav2.1), blocked by agatoxins; the N-type (Cav 2.2), blocked by conotoxins; and L-type (Cav1.1-4), blocked by dihydropyridines. From unpublished experiments, we have determined that two calcium channels, Cav1.2 and Cav2.2, are expressed at a detectable level in the naïve PC12 cells used in this study. Quantitative RTPCR showed that the expression of Cav1.2 but not expression of Cav2.2 was reduced in the MKP3-expressing cells. The reduced levels of Cav1.2 expression would explain the absence of neurotransmitter release, but does not exclude the possibility that other components of the release process could be affected by MKP3 expression. To determine whether the downregulation of Cav1.2 expression was the only process responsible for the absent neurotransmitter release, Cav1.2 or Cav2.2 was reintroduced into the MKP3-expressing cells, and this concluding experiment established that the reintroduction of Cav1.2 but not Cav2.2 was sufficient to recover the dopamine release in MKP3-expressing PC12 cells. I can therefore conclude that MKP3 reduces neurotransmitter release through a downregulation of Cav1.2 expression. 4.2. Map kinases in calcium channel regulation

4. Discussion 4.1. MKP3 attenuates neurotransmitter release through downregulation of the expression of an L-type calcium channel In this study we explored the involvement of MKPs, endogenous inactivators and regulators of MAPK signaling, in vesicular release of monoamines in rat PC12 cells. PC12 cells are derived from a pheochromocytoma and display characteristics of chromaffin and sympathetic cell precursors [26]. MKPs are dual-specificity phosphatases that inactivate and dephosphorylate MAPKs, including ERK1/2, p38, and JNK MAPKs [11]. Based on sequence similarity, substrate specificity, and subcellular localization, MKPs can be subdivided into distinct groups. One group consists of DUSP1/MKP1, DUSP2/PAC-1, DUSP4/MKP2, and DUSP5/hVH-3. These proteins are encoded by highly inducible genes, which are rapidly upregulated in response to both mitogenic and/or stress stimuli at the transcriptional level. They are all nuclear proteins and with the exception of DUSP5/hVH-3, which appears to specifically target the ERK1/2 MAPKs, they exhibit broad substrate specificity and can inactivate ERK, JNK, and p38 MAPKs. The second group of MKPs consists of DUSP6/MKP3, DUSP7/MKPX, and DUSP9/MKP4. These three proteins are cytoplasmic enzymes and exhibit a degree of selectivity toward the ERK1/2 MAPKs. DUSP6/MKP3 in particular has little or no activity toward either JNK or p38 when overexpressed in mammalian cells. In the present study, we employed MKP1 and MKP3 as representative members of each of these two groups and explored their role in neurotransmitter release. Significantly, initial experiments (Fig. 1) revealed an important role of MKP3 but not MKP1 in the release process as only MKP3 expression reduced dopamine

MKP3 mediates its effects on gene expression through the dephosphorylation of the ERK1/2 MAPKs, and interestingly it has already been shown that extracellular activators of MAPK signaling such as growth factors can affect acute and long-term regulation of Cav1.2 channels. Studies of the effect of growth factors such as epidermal growth factor (EGF), NGF, glial cell-derived neurotrophic factor (GDNF), and BDNF have shown that all these factors can affect both gene expression and activity of the calcium channels. The effects appear to be cell type dependent as they are not always in agreement. Most studies do agree with the results presented here that activating MAPK signaling will upregulate calcium channel currents. In embryonic forebrain neurons, it was reported that 4–6 days of exposure to NGF but not to BDNF would increase L-type and N-type calcium currents [27]. In hippocampal neurons, NGF upregulated L-type calcium channel currents in an ERK1/2-dependent manner and BDNF upregulated other non-L-type calcium channels [28]. In rat dorsal root ganglion neurons, tonic EGF, NGF, and GDNF maintain the expression of calcium channels and that NGF and GDNF can acutely activate the same channels in an ERK1/2dependent manner [29,30]. L-type channels appeared to be more affected by tonic NGF levels [30]. The acute upregulation of calcium channel activity in the above study is opposite to the effect observed in our study where we observe no effect of acute inhibition of ERK1/2 activity on neurotransmitter release (Fig. 2A) and it is also opposite to a recent study of mouse cortical neurons linking L-type calcium channel activity to exocytosis, as this study found that the inhibition of ERK1/2 activity resulted in an upregulation in L-type calcium current, causing increased vesicle exocytosis [10]. I speculate the reason for these discrepancies could be attributed

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

to the fact that all three studies were carried out in different cell systems and therefore also suggest that there must be cell type specific regulation of MAPK mediated regulation of calcium channel function. In further support for MAPK dependent regulation of calcium channel expression, genetic studies expressing the upstream ERK1/2 activator Ras have also found MAPK-dependent effects on calcium channel currents. A study in PC12 cells showed that an NGF-induced increase in calcium currents was absent in cells expressing a dominant-negative version of Ras [31]. Interestingly, in the same study using calcium current recordings and in another study using calcium uptake [32], investigators found that the Ras dominant-negative mutant expressing PC12 cells maintained calcium channel expression. MKP3 expression therefore appears to have a stronger effect on ERK1/2 dependent regulation of cav1.2 expression than the mutant Ras. 4.3. Physiological role of MKP3 mediated regulation of neurotransmitter release Expression and activity of MKP3 is tightly regulated as MKP3 is sensitive to proteasomal degradation in a phosphorylationdependent manner [33] and is expressed in very low basal amounts in the brain [34]. Particularly interesting is the fact that MKP3 is upregulated by neuronal activity [12,13,34], and by methamphetamine treatment that pharmacologically enhances neuronal activity in rats [35]. With the results presented here showing decreased neurotransmitter release when MKP3 expression is increased makes us speculate that the physiological relevance of MKP3 expression and its regulation of Cav1.2 expression is to function as an important regulator of neural plasticity within the presynaptic neuron. Possibly the upregulation of MKP3 expression following increased neuronal activity compensates and decreases the elevated neuronal activity by regulating the expression of genes such as Cav1.2, reported here, and ultimately decreases neurotransmitter release. Interestingly, we previously showed that MKP3 stabilizes the surface expression of the dopamine transporter that is responsible for the reuptake of dopamine into the presynaptic dopaminergic neuron [36] showing another way upregulation of MKP3 expression can affect neuronal activity by affecting extracellular levels of neurotransmitter. 4.4. Genetic studies implicates MKP3 and Cav1.2 in psychiatric and cardiovascular diseases The possible implications of MKP3 in regulating Cav1.2 expression and neurotransmitter homeostasis are further highlighted by genetic studies that have linked both Cav1.2 and MKP3 to psychiatric diseases. One of the most consistent findings from genetic association studies of patients with bipolar disorders, schizophrenia, or major depression is mutations in the CACNA1C gene encoding Cav1.2 [37]. In addition, a mutation in this gene causes Timothy’s syndrome [38]. This disease presents with arrhythmias, congenital heart disease, and autism spectrum symptoms. MKP3, conversely, has been found to be significantly associated with bipolar disorder in a cohort of Korean women [39] and to affect in vitro effects of lithium, a commonly used treatment of bipolar disorder [40]. Further supporting a role for MKP3 in mood disorders, postmortem studies have found an association with the expression levels of MKP3 in frontal cortices of patients with bipolar disorders, particularly in women [41,42]. The sex-specific effect of MKP3 is interesting in light of an animal study of heterozygous Cav1.2 knockout mice, expressing decreased Cav1.2 protein levels. These animals display decreased L-type calcium channel current as adults and are protected against depression-like phenotypes [43]. Significantly, this finding, although reported in both sexes, was more robust in females. This study together with the genetic

229

studies of Korean women and the postmortem studies of MKP3 expression could suggest a sex-specific functional link between MKP3-regulated expression of Cav1.2, as shown in this study, and depressive phenotypes. The possible significance of the interaction between MKP3 and calcium channels reported in this study is not limited to the CNS. For example, L-type channels including Cav1.2 play an important role in the heart and are responsible for the entry of calcium during the upstroke of the action potential that is involved in depolarizing the cardiac nodal cells. Calcium entry through the L-type calcium channels is also essential for the activation of the contraction of the cardiac muscle [44]. Therefore it is interesting that one of the main findings from studies of the physiological function of MKP3 is phenotypes involving the heart. MKP3 knockout mice have enlarged hearts [45], and a zebra fish study found that a specific inhibitor of MKP3 function would affect heart size [46]. Whether the effects of MKP3 on heart anatomy involve effects on calcium channel expression and function remains to be established. 5. Conclusion In conclusion, I have in this study identified a role of the MAPK phosphatase MKP3 in neurotransmitter release. Specifically I have shown that MKP3 downregulates neurotransmitter release from large dense core vesicles in rat pheochromocytoma PC12 cells. I have established that the mechanism through which MKP3 attenuates exocytosis is via a downregulation of the expression of voltage-dependent calcium channel Cav1.2 expression. Finally, because MKP3 function is regulated by neuronal activity I speculate that this process is important for neurotransmitter homeostasis in situations of aberrant neuronal activity. Acknowledgment I am grateful to Susan G. Amara for scientific and financial support during the initial development of this project. I am also indebted to Mee H. Choi for valuable discussions, helpful comments, and for technical assistance with the calcium imaging experiments. Finally, I also thank Andreia C.K. Fontana and Diana M. Winters for helpful comments on the manuscript. References [1] S.M. Wojcik, N. Brose, Regulation of membrane fusion in synaptic excitationsecretion coupling: speed and accuracy matter, Neuron 55 (2007) 11–24. [2] C.L. Waites, C.C. Garner, Presynaptic function in health and disease, Trends in Neurosciences 34 (2011) 326–337. [3] T.C. Sudhof, The synaptic vesicle cycle, Annual Review of Neuroscience 27 (2004) 509–547. [4] W.E. Ghijsen, A.G. Leenders, Differential signaling in presynaptic neurotransmitter release, Cellular and Molecular Life Sciences 62 (2005) 937–954. [5] A.G. Leenders, Z.H. Sheng, Modulation of neurotransmitter release by the second messenger-activated protein kinases: implications for presynaptic plasticity, Pharmacology and Therapeutics 105 (2005) 69–84. [6] W.A. Catterall, A.P. Few, Calcium channel regulation and presynaptic plasticity, Neuron 59 (2008) 882–901. [7] M.C. Lawrence, A. Jivan, C. Shao, L. Duan, D. Goad, E. Zaganjor, J. Osborne, K. McGlynn, S. Stippec, S. Earnest, W. Chen, M.H. Cobb, The roles of MAPKs in disease, Cell Research 18 (2008) 436–442. [8] J.N. Jovanovic, A.J. Czernik, A.A. Fienberg, P. Greengard, T.S. Sihra, Synapsins as mediators of BDNF-enhanced neurotransmitter release, Nature Neuroscience 3 (2000) 323–329. [9] S.A. Kushner, Y. Elgersma, G.G. Murphy, D. Jaarsma, G.M. van Woerden, M.R. Hojjati, Y. Cui, J.C. LeBoutillier, D.F. Marrone, E.S. Choi, C.I. De Zeeuw, T.L. Petit, L. Pozzo-Miller, A.J. Silva, Modulation of presynaptic plasticity and learning by the H-ras/extracellular signal-regulated kinase/synapsin I signaling pathway, Journal of Neuroscience 25 (2005) 9721–9734. [10] J. Subramanian, A. Morozov, Erk1/2 inhibit synaptic vesicle exocytosis through L-type calcium channels, Journal of Neuroscience 31 (2011) 4755–4764. [11] R.J. Dickinson, S.M. Keyse, Diverse physiological functions for dual-specificity MAP kinase phosphatases, Journal of Cell Science 119 (2006) 4607–4615.

230

O.V. Mortensen / Cell Calcium 53 (2013) 224–230

[12] M. Kodama, D.S. Russell, R.S. Duman, Electroconvulsive seizures increase the expression of MAP kinase phosphatases in limbic regions of rat brain, Neuropsychopharmacology 30 (2005) 360–371. [13] G. Xiang, L. Pan, W. Xing, L. Zhang, L. Huang, J. Yu, R. Zhang, J. Wu, J. Cheng, Y. Zhou, Identification of activity-dependent gene expression profiles reveals specific subsets of genes induced by different routes of Ca(2+) entry in cultured rat cortical neurons, Journal of Cellular Physiology 212 (2007) 126–136. [14] T.D. Helton, W. Xu, D. Lipscombe, Neuronal L-type calcium channels open quickly and are inhibited slowly, Journal of Neuroscience 25 (2005) 10247–10251. [15] J.Q. Pan, D. Lipscombe, Alternative splicing in the cytoplasmic II-III loop of the N-type Ca channel alpha 1B subunit: functional differences are beta subunitspecific, Journal of Neuroscience 20 (2000) 4769–4775. [16] Y.A. Ushkaryov, A. Rohou, S. Sugita, alpha-Latrotoxin and its receptors, Handbook of Experimental Pharmacology 184 (2008) 171–206. [17] P. Mandela, G.A. Ordway, The norepinephrine transporter and its regulation, Journal of Neurochemistry 97 (2006) 310–333. [18] O.V. Mortensen, S.G. Amara, Gain of function mutants reveal sites important for the interaction of the atypical inhibitors benztropine and bupropion with monoamine transporters, Journal of Neurochemistry 98 (2006) 1531–1540. [19] P. Mandela, G.A. Ordway, KCl stimulation increases norepinephrine transporter function in PC12 cells, Journal of Neurochemistry 98 (2006) 1521–1530. [20] D.B. Pereira, A.P. Carvalho, C.B. Duarte, Non-specific effects of the MEK inhibitors PD098,059 and U0126 on glutamate release from hippocampal synaptosomes, Neuropharmacology 42 (2002) 9–19. [21] J.S. Schonn, C. Desnos, J.P. Henry, F. Darchen, Transmitter uptake and release in PC12 cells overexpressing plasma membrane monoamine transporters, Journal of Neurochemistry 84 (2003) 669–677. [22] M.M. Usowicz, H. Porzig, C. Becker, H. Reuter, Differential expression by nerve growth factor of two types of Ca2+ channels in rat phaeochromocytoma cell lines, Journal of Physiology 426 (1990) 95–116. [23] H. Liu, R. Felix, C.A. Gurnett, M. De Waard, D.R. Witcher, K.P. Campbell, Expression and subunit interaction of voltage-dependent Ca2+ channels in PC12 cells, Journal of Neuroscience 16 (1996) 7557–7565. [24] D. Janigro, G. Maccaferri, J. Meldolesi, Calcium channels in undifferentiated PC12 rat pheochromocytoma cells, FEBS Letters 255 (1989) 398–400. [25] S.C. Taylor, C. Peers, Store-operated Ca2+ influx and voltage-gated Ca2+ channels coupled to exocytosis in pheochromocytoma (PC12) cells, Journal of Neurochemistry 73 (1999) 874–880. [26] D. Vaudry, P.J. Stork, P. Lazarovici, L.E. Eiden, Signaling pathways for PC12 cell differentiation: making the right connections, Science 296 (2002) 1648–1649. [27] E.S. Levine, C.F. Dreyfus, I.B. Black, M.R. Plummer, Differential effects of NGF and BDNF on voltage-gated calcium currents in embryonic basal forebrain neurons, Journal of Neuroscience 15 (1995) 3084–3091. [28] P. Baldelli, P.E. Forni, E. Carbone, BDNF, NT-3 and NGF induce distinct new Ca2+ channel synthesis in developing hippocampal neurons, European Journal of Neuroscience 12 (2000) 4017–4032. [29] E.M. Fitzgerald, Regulation of voltage-dependent calcium channels in rat sensory neurones involves a Ras-mitogen-activated protein kinase pathway, Journal of Physiology 527 (Pt 3) (2000) 433–444. [30] A.J. Woodall, M.A. Richards, D.J. Turner, E.M. Fitzgerald, Growth factors differentially regulate neuronal Cav channels via ERK-dependent signalling, Cell Calcium 43 (2008) 562–575. [31] J.D. Pollock, S.G. Rane, p21ras signaling is necessary but not sufficient to mediate neurotrophin induction of calcium channels in PC12 cells, Journal of Biological Chemistry 271 (1996) 8008–8014.

[32] E. Bloch-Shilderman, H. Jiang, S. Abu-Raya, M. Linial, P. Lazarovici, Involvement of extracellular signal-regulated kinase (ERK) in pardaxin-induced dopamine release from PC12 cells, Journal of Pharmacology and Experimental Therapeutics 296 (2001) 704–711. [33] S. Marchetti, C. Gimond, J.C. Chambard, T. Touboul, D. Roux, J. Pouyssegur, G. Pages, Extracellular signal-regulated kinases phosphorylate mitogen-activated protein kinase phosphatase 3/DUSP6 at serines 159 and 197, two sites critical for its proteasomal degradation, Molecular and Cellular Biology 25 (2005) 854–864. [34] U. Boschert, R. Dickinson, M. Muda, M. Camps, S. Arkinstall, Regulated expression of dual specificity protein phosphatases in rat brain, Neuroreport 9 (1998) 4081–4086. [35] M. Takaki, H. Ujike, M. Kodama, Y. Takehisa, K. Nakata, S. Kuroda, Two kinds of mitogen-activated protein kinase phosphatases, MKP-1 and MKP-3, are differentially activated by acute and chronic methamphetamine treatment in the rat brain, Journal of Neurochemistry 79 (2001) 679–688. [36] O.V. Mortensen, M.B. Larsen, B.M. Prasad, S.G. Amara, Genetic complementation screen identifies a mitogen-activated protein kinase phosphatase, MKP3, as a regulator of dopamine transporter trafficking, Molecular Biology of the Cell 19 (2008) 2818–2829. [37] S. Bhat, D.T. Dao, C.E. Terrillion, M. Arad, R.J. Smith, N.M. Soldatov, T.D. Gould, CACNA1C (Ca(v)1.2) in the pathophysiology of psychiatric disease, Progress in Neurobiology 99 (2012) 1–14. [38] I. Splawski, K.W. Timothy, L.M. Sharpe, N. Decher, P. Kumar, R. Bloise, C. Napolitano, P.J. Schwartz, R.M. Joseph, K. Condouris, H. Tager-Flusberg, S.G. Priori, M.C. Sanguinetti, M.T. Keating, Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism, Cell 119 (2004) 19–31. [39] K.Y. Lee, Y.M. Ahn, E.J. Joo, J.S. Chang, Y.S. Kim, The association of DUSP6 gene with schizophrenia and bipolar disorder: its possible role in the development of bipolar disorder, Molecular Psychiatry 11 (2006) 425–426. [40] S.H. Kim, S.Y. Shin, K.Y. Lee, E.J. Joo, J.Y. Song, Y.M. Ahn, Y.H. Lee, Y.S. Kim, The genetic association of DUSP6 with bipolar disorder and its effect on ERK activity, Progress in Neuro-Psychopharmacology and Biological Psychiatry 37 (2012) 41–49. [41] B.W. Higgs, M. Elashoff, S. Richman, B. Barci, An online database for brain disease research, BMC Genomics 7 (2006) 70. [42] J. Struyf, S. Dobrin, D. Page, Combining gene expression, demographic and clinical data in modeling disease: a case study of bipolar disorder and schizophrenia, BMC Genomics 9 (2008) 531. [43] D.T. Dao, P.B. Mahon, X. Cai, C.E. Kovacsics, R.A. Blackwell, M. Arad, J. Shi, P.P. Zandi, P. O’Donnell, J.A. Knowles, M.M. Weissman, W. Coryell, W.A. Scheftner, W.B. Lawson, D.F. Levinson, S.M. Thompson, J.B. Potash, T.D. Gould, Mood disorder susceptibility gene CACNA1C modifies mood-related behaviors in mice and interacts with sex to influence behavior in mice and diagnosis in humans, Biological Psychiatry 68 (2010) 801–810. [44] I. Bodi, G. Mikala, S.E. Koch, S.A. Akhter, A. Schwartz, The L-type calcium channel in the heart: the beat goes on, Journal of Clinical Investigation 115 (2005) 3306–3317. [45] M. Maillet, N.H. Purcell, M.A. Sargent, A.J. York, O.F. Bueno, J.D. Molkentin, DUSP6 (MKP3) null mice show enhanced ERK1/2 phosphorylation at baseline and increased myocyte proliferation in the heart affecting disease susceptibility, Journal of Biological Chemistry 283 (2008) 31246–31255. [46] G. Molina, A. Vogt, A. Bakan, W. Dai, P. Queiroz de Oliveira, W. Znosko, T.E. Smithgall, I. Bahar, J.S. Lazo, B.W. Day, M. Tsang, Zebrafish chemical screening reveals an inhibitor of Dusp6 that expands cardiac cell lineages, Nature Chemical Biology 5 (2009) 680–687.