cJun N-terminal kinase (JNK) mediates cortico-striatal signaling in a model of Parkinson's disease

The cJun N-terminal kinase (JNK) signaling pathway has been extensively studied with regard to its involvement in neurodegenerative processes, but little is known about its functions in neurotransmission. In a mouse model of Parkinson’s disease (PD), we show that the pharmacological activation of dopamine D1 receptors (D1R) produces a large increase in JNK phosphorylation. This effect is secondary to dopamine depletion, and is restricted to the striatal projection neurons that innervate directly the output structures of the basal ganglia (dSPN). Activation of JNK in dSPN relies on cAMP-induced phosphorylation of the dopamine- and cAMP-regulated phosphoprotein of 32 kDa (DARPP-32), but does not require N-methyl-D-aspartate (NMDA) receptor transmission. Electrophysiological experiments on acute brain slices from PD mice show that inhibition of JNK signaling in dSPN prevents the increase in synaptic strength caused by activation of D1Rs. Together, our findings show that dopamine depletion confers to JNK the ability to mediate dopamine transmission, informing the future development of therapies for PD. we investigated whether striatal JNK mediates the cellular and synaptic effects of D1R activation in naïve mice and in the 6-hydroxydopamine (6-OHDA) mouse model of PD. We find that activation of D1Rs with a specific agonist or L-DOPA increases the phosphorylation of JNK and cJun in SPNs that project directly to the output nuclei of the basal ganglia (dSPN). This effect is observed in concomitance with dopamine depletion, which is one of the cardinal features of PD, and is mediated by cAMP-dependent phosphorylation of dopamine- and cAMP-regulated phosphoprotein of 32 kDa (DARPP-32). Finally, we show that JNK is involved in the D1R-mediated regulation of synaptic mechanisms of plasticity at cortico-striatal synapses on dSPN in the disease state, but not under physiological conditions. to a role JNK the striatal synaptic plasticity. this regulation is absent in naïve mice and occurs only in a PD model, characterized by the loss of dopaminergic input to the

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Introduction
Dopamine transmission is involved in numerous basic physiological processes and its abnormal activity is characteristic of several neuropsychiatric and neurodegenerative disorders, including drug addiction and Parkinson's disease (PD). Dopamine's effects are enacted through binding to two major types of G-protein coupled receptors, the D1 (D1R) and D2 (D2R) receptors, which are linked to stimulation and inhibition of cAMP signaling, respectively [1].
In experimental models of PD, the loss of dopaminergic afferents to the striatum is associated with sensitized dopamine transmission. Previous work shows that this sensitization is particularly prominent for D1Rs [2,3], which are selectively expressed in the striatal projection neurons (SPNs) that form the direct pathway to the output nuclei of the basal ganglia [4]. D1R sensitization in response to dopamine depletion has been attributed to abnormal recruitment of D1Rs at the cell surface, as well as increased levels of G olf and adenylyl cyclase [5][6][7][8]. These changes dramatically enhance the effects produced by dopaminergic drugs on cAMP signaling [9]. In addition, sensitized D1R transmission confers on L-DOPA, the most commonly prescribed medication for PD, the ability to activate non-canonical signal transduction pathways, such as the extracellular signal-regulated kinases (ERK) and the mammalian target of rapamycin [9][10][11][12][13][14]. Whether activation of sensitized D1Rs also leads to the stimulation of other downstream targets in SPNs remains to be established.
The cJun N-terminal kinase (JNK) signaling pathway is a member of the mitogen-activated protein kinase family, which has been extensively studied with regard to cell survival and proliferation. The primary target of JNK is the early response transcription factor, cJun, which is a component of the adaptor protein-1 (AP-1) [15,16]. JNK-mediated phosphorylation enhances the transcriptional efficiency of cJun by increasing its binding to DNA at promoter sites [17,18]. Work in human neuroblastoma cells showed that D1Rs activate JNK via stimulation of cAMP-dependent protein kinase [19]. Moreover, To address this question, we investigated whether striatal JNK mediates the cellular and synaptic effects of D1R activation in naïve mice and in the 6-hydroxydopamine (6-OHDA) mouse model of PD.
We find that activation of D1Rs with a specific agonist or L-DOPA increases the phosphorylation of JNK and cJun in SPNs that project directly to the output nuclei of the basal ganglia (dSPN). This effect is observed in concomitance with dopamine depletion, which is one of the cardinal features of PD, and is mediated by cAMP-dependent phosphorylation of dopamine-and cAMP-regulated phosphoprotein of 32 kDa (DARPP-32). Finally, we show that JNK is involved in the D1R-mediated regulation of synaptic mechanisms of plasticity at cortico-striatal synapses on dSPN in the disease state, but not under physiological conditions.
A C C E P T E D M A N U S C R I P T 6

6-OHDA lesion
In order to deplete dopamine in the right striatum, mice were injected with 6-hydroxydopamine-HCl (6-OHDA; Sigma-Aldrich, Sweden AB) into the ipsilateral medial forebrain bundle (MFB). All animals received subcutaneous injections of Temgesic ® as analgesic at a dose of 0.1 mg/kg, before the surgical procedure. Anesthesia was induced with 4% isofluorane and maintained with 2% isoflurane.

ACCEPTED MANUSCRIPT
A C C E P T E D M A N U S C R I P T 7 conjugated secondary antibody (1:400, Jackson Laboratories). In a subset of experiments, phospho-cJun was also visualized using an antibody against phospho-Ser63-cJun with undistinguishable results (data not shown). A primary antibody against EGFP (1:1000; Aves labs, Tigard, OR, USA) and a 488conjugated secondary antibody (1:500; Life Technologies, Paisley, UK) were used to visualize EGFP.
Triple-labeling for P-JNK/P-cJun/EGFP was performed as described in [23]. Briefly, P-JNK staining was carried out using the TSA technique and microwave treatment of the sections was performed before proceeding with the P-cJun/EGFP staining. Immunofluorescence against TH was carried out using an anti-TH primary antibody (1:500, AB152, Millipore) and a Cy3-conjugated secondary antibody (1:400, Jackson Laboratories). Single-, double-and triple-labeled images of the dorsal striatum were captured using sequential laser scanning confocal microscopy (Zeiss LSM, Carl Zeiss, Jena, Germany) at 20, 40 and 63X magnification. Quantification of staining was obtained averaging two 20X images per section, in two sections per animal, by counting positive cells above background using Image J software and the average number of positive cells was reported.

Ex vivo slice experiments
Mice with a unilateral injection of 6-OHDA in the MFB were killed by decapitation and the brains rapidly removed. Coronal slices (250 μm) were prepared using a vibratome (Leica, Germany) and punches corresponding to the dorsal striatum were dissected out from the control, or the 6-OHDA-lesion side. Two punches were placed in individual 5-ml polypropylene tubes containing 2 ml of Krebs-Ring

ACCEPTED MANUSCRIPT
A C C E P T E D M A N U S C R I P T

Electrophysiology
Whole-cell recordings were performed on horizontal brain slices, which preserve intact corticostriatal connections [24]. SPN in the DLS were visualized under infrared-differential interference contrast and were identified on the basis of morphological and electrical properties [25]. After recording, the identity

ACCEPTED MANUSCRIPT
A C C E P T E D M A N U S C R I P T

Statistics
Biochemical data were analyzed with unpaired, two-tailed t-test, one-way (1W) or two-way (2W) analysis of variance (ANOVA), where lesion and treatment/time/genotype were the independent variables. Electrophysiological data were analyzed by one-way repeated measures (1W RM) ANOVA for comparisons within a group, and by 1W ANOVA for between-group comparisons (GraphPad Instat 6

A C C E P T E D M A N U S C R I P T
10 software). Post hoc analyses (Tukey and Dunnet multiple comparison tests) were only performed for ANOVAs that yielded significant main effects. For comparisons of two groups, an independent twosample t-test was used.

L-DOPA regulates JNK signaling in the D1R expressing dSPN
To determine whether the effects of L-DOPA were cell-type specific, we examined changes in phosphorylation of JNK and cJun in dSPN, which express D1Rs, and in the iSPNs, which are instead enriched in D2Rs. For this purpose, 6-OHDA was injected in mice expressing EGFP in dSPN (Drd1a-

A C C E P T E D M A N U S C R I P T
11 EGFP mice), or iSPNs (Drd2-EGFP mice). Mice were then treated with L-DOPA (10 mg/kg, i.p.) and perfused for immunohistochemical analysis 30 min later. In Drd2-EGFP mice, L-DOPA-induced phosphorylation of JNK ( Fig. 2A, upper panels) and cJun (Fig. 2B, upper panels) was localized in EGFPnegative cells (p = 0.0002 for P-JNK and p = 0.0007 for P-cJun). Conversely, in Drd1a-EGFP mice L-DOPA-induced phosphorylation of JNK ( Fig. 2A, lower panels) and cJun (Fig. 2B, lower panels) was largely co-localized with EGFP (p = 0.0005 for P-JNK and p = 0.0012 for P-cJun). High magnification of the striatum from a Drd2-EGFP mouse revealed that P-JNK and P-cJun were increased in the same EGFP-negative SPNs (Fig. 2C). These results indicate that L-DOPA enhances the phosphorylation of JNK and cJun specifically in dSPN.

Pharmacological activation of D1Rs promotes JNK signaling in the dopamine depleted striatum
The

A C C E P T E D M A N U S C R I P T
14 mediated synaptic effects (naive + SKF/JNK-IN-8, 90 ± 8 % of baseline, n = 6, p > 0.05; Fig. 6C). This is consistent with the lack of D1R-mediated JNK activation in the intact striatum ( Fig. 3E-H). After recordings, dSPN cell identity was confirmed by immunostaining for substance P (positive) and A2A (negative) (data not shown) [25] Together, these results demonstrate that dopamine depletion confers the ability to D1R to signal by JNK, supporting a role for the dopaminergic regulation of the JNK signaling pathway in striatal dysfunction.

A C C E P T E D M A N U S C R I P T
15

Discussion
We have identified the JNK pathway as a signaling mechanism implicated in dopamine transmission, and provide evidence in an experimental model of PD that JNK is required for D1R-dependent modulation of corticostriatal synaptic plasticity. In

A C C E P T E D M A N U S C R I P T
16 during acute inflammation [38]. The present data suggest that, in the brain, PP-1 is an additional phosphatase implicated in dopamine-mediated control of JNK signaling.
The effect of DARPP-32 inactivation on the phosphorylation of cJun is less prominent than that observed on JNK. This suggests that following D1R sensitization, abnormal activation of PKA, which is upstream of DARPP-32, may be sufficient to increase cJun phosphorylation independently of parallel inhibition of PP-1. It is also possible that the residual phosphorylation of JNK and Ras/ERK [9] still observed after genetic inactivation of DARPP-32 may concur to maintain cJun in a partially active state.
Future studies will be needed to examine these possibilities.
Stimulation of mitogen-activated protein kinases typically depends on NMDA receptor transmission.