Dopamine D4 Receptor Is a Regulator of Morphine-Induced Plasticity in the Rat Dorsal Striatum

Long-term exposition to morphine elicits structural and synaptic plasticity in reward-related regions of the brain, playing a critical role in addiction. However, morphine-induced neuroadaptations in the dorsal striatum have been poorly studied despite its key function in drug-related habit learning. Here, we show that prolonged treatment with morphine triggered the retraction of the dendritic arbor and the loss of dendritic spines in the dorsal striatal projection neurons (MSNs). In an attempt to extend previous findings, we also explored whether the dopamine D4 receptor (D4R) could modulate striatal morphine-induced plasticity. The combined treatment of morphine with the D4R agonist PD168,077 produced an expansion of the MSNs dendritic arbors and restored dendritic spine density. At the electrophysiological level, PD168,077 in combination with morphine altered the electrical properties of the MSNs and decreased their excitability. Finally, results from the sustantia nigra showed that PD168,077 counteracted morphine-induced upregulation of μ opioid receptors (MOR) in striatonigral projections and downregulation of G protein-gated inward rectifier K+ channels (GIRK1 and GIRK2) in dopaminergic cells. The present results highlight the key function of D4R modulating morphine-induced plasticity in the dorsal striatum. Thus, D4R could represent a valuable pharmacological target for the safety use of morphine in pain management.


Introduction
Long-lasting use of drugs of abuse elicits persistent molecular and cellular neuroadaptive changes within discrete brain regions-e.g., ventral tegmental area (VTA), ventral and dorsal striatum, prefrontal cortex, amygdala, or hippocampus-which have been identified as the neural substrate for behavioral abnormalities driving addiction, drug craving, and relapse [1]. Such drug-mediated alterations affect several neurotransmitters systems-e.g., dopaminergic, GABAergic or glutamatergic systems-signal transduction pathways, neuronal activity, cellular architecture remodeling and synaptic strength [2][3][4][5][6]. Although there is an extensive amount of data regarding drug-mediated modifications, yet it is a major challenge to understand how this wide variety of neuroadaptations converge and operate together in the different stages of the addiction cycle-binge/intoxication', 'withdrawal/negative affect', and 'preoccupation/anticipation' [1].
Most of the studies in drug-induced neuroadaptations have preferentially focused in the traditional areas of the reward system-i.e., mesolimbic dopaminergic pathway arising from the VTA which provides projections into the nucleus accumbens (NAc), the major component of the ventral striatum. However, nowadays major attention is also being given to the caudate putamen (CPu)-dorsal striatum-since this region play a pivotal role in goal-directed behavior, drug-related habit learning and automaticity of drug consumption [7][8][9]. Therefore, it has been hypothesized that the transition from recreational drug use to compulsive drug abuse, and thus the consolidation of drug-related instrumental behaviors, comprises a transition from the ventral to the dorsal areas of the striatum activity [10].
Morphine is one of the most powerful analgesic drugs used to relieve pain though highly addictive, making difficult to design a proper prescription schedule. Indeed, longterm morphine consumption promotes neurobiological adaptations including synaptic and structural plasticity in some brain regions, which ultimately contribute to the development of addiction [5,11]. So far, it has been reported that morphine produces neuronal morphological changes, such as alterations of the dendritic arbor complexity and dendritic spines at nucleus accumbens (NAc) medium spiny neurons (MSNs) and at cortical and hippocampal pyramidal cells [2,5,[12][13][14][15]. Chronic morphine treatment also produces neuro-adaptative changes in dopamine neurons of the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc), which correlates with altered neuron excitability and dopamine efflux [16][17][18]. Though not as extensively investigated as the NAc, the CPu also undergoes a morphine-induced dysregulation of nigral dopamine inputs and signaling [16,[19][20][21]. However, there is no evidence on whether this drug potentially produces morphological plasticity on dorsal striatal MSNs similar to those described in other brain regions as described above.
The identification of novel strategies to suppress the addictive properties of morphine has emerged as a hot topic in the current investigation in opioid addiction [22]. Several studies have highlighted the dopamine D 4 receptor (D 4 R) as a key modulator in the addicted behavior, since the deficiency in this receptor seems to increase the susceptibility to the development of drugs abuse [23,24]. Of high interest is the evidence that the selective stimulation of the D 4 R decreases morphine-induced hyperlocomotion, reward and withdrawal syndrome, without interfering with the analgesic properties of morphine [16]. These results were strengthened by the demonstration that D 4 R stimulation counteracts morphine-induced activation of the nigrostriatal pathway leading to a restoring of dorsal striatal dopamine tone, and also by the ability of the D 4 R to modulate molecular alterations and long-term µ opioid receptor (MOR) sensitization in the CPu [16,[25][26][27]. Despite these evidences, the exact molecular mechanisms by which D 4 R appears to prevent the addictive effects of morphine are not yet fully understood, but an antagonistic D 4 R-MOR interaction in the CPu might occur through the formation of heteroreceptor complexes [16,28,29].
In this perspective, the aim of the present study was to determine how the continuous exposure to morphine would induce structural plasticity and changes in the intrinsic membrane excitability and firing properties of the dorsal striatal MSNs, and whether these neuroadaptations were modulated by D 4 R activation.

Animals
Experiments were conducted in 1-2 month-old male Sprague-Dawley rats (Charles River, Barcelona, Spain). The animals were maintained on a standard light/dark cycle (12/12 h), constant room temperature (20 ± 2 • C) and relative humidity (65 ± 75%) with food and water available ad libitum. Animal care and procedures were conducted in accordance with protocols approved by the Ethical Committee of the University of Málaga (CEUMA 79-2019-A) and the University of Bordeaux (CEEA-50), guidelines from the European Union Council Directive 86/609/EEC, as well as the Spanish Government (R.D. 53/2013).

Drug Administration
Morphine sulfate was obtained from Alcaliber S.A. (Madrid, Spain) subsequent to receiving authorization from Spanish Agency of Medicines and Medical Devices (Spanish Government). PD168,077 maleate was supplied by Tocris Bioscience (Avonmouth, UK). The specificity of PD168,077 as a D 4 R agonist has been extensively proved in prior works using the highly selective antagonist L745,870 [16,[25][26][27]. All drugs were dissolved in the vehicle solution consistent in 2% dimethyl sulfoxide (DMSO) and 0.9% NaCl. We have demonstrated in a previous report that this amount of DMSO had no effect on receptors function [30].
Rats received 7 or 14 days of continuous administration of vehicle, morphine (20 mg/kg/d) and/or PD168,077 (1 mg/kg/d) by an osmotic pump (Alzet ® osmotic pumps, Cupertino, CA, USA) that was subcutaneously implanted under deep anesthesia (75 mg/kg ketamine and 0.5 mg/kg medetomidine, i.p.) between the shoulder blades. Skin was sutured with sterile non absorbable suture and the surgical site was disinfected with a topical antiseptic (povidone iodine 10% solution). During the surgery and recovery, animals were kept warn using a heating pad. Animals were sacrificed the last day of the continuous administration of drugs.

Whole-Cell Current-Clamp Recordings
Each section was transferred into a recording chamber and was continuously perfused with an oxygenated ACSF at a rate of 3.5 mL/min. MSNs within the dorsolateral part of the CPu were identified by their morphological characteristics under infrared differential interfere contrast (IR-DIC) optics (Zeiss examiner Z.1).
The patch pipette was filled with a KGluconate-based solution containing (in mM): 140 KGluconate, 3.8 NaCl, 1 MgCl 2 , 10 HEPES, 0.1 Na 4 EGTA, 2 Mg, 1.5 ATP, and 0.4 Na 3 GTP. The day of the experiment the solution was supplemented by biocytin (2.5-5 mg/mL) and by Alexa fluor 488 (2 µM) to allow morphological analysis of the recorded cells. The pH and osmolarity of the pipette solution were 7.3 and 290 mOsm, respectively. The junction potential between the electrode solution and the external media (empirically estimated as 13 mV) was not corrected. Electrode signals were low-pass filtered at 4 kHz and sampled at 20 kHz. Only one neuron was recorded per slice to avoid uncertainty in the reconstruction phase.
The recordings usually begun in the voltage clamp mode. The neurons were maintained at −80 mV, and 5 successive voltage steps of −10 mV were applied. The average current response was analyzed off-line and cell capacity (Cm), membrane resistance (Rm) and access resistance (Ra) were calculated. In some neurons, only current clamp was used, and these three parameters were calculated from the voltage response to a −25pA stimulation. In the current clamp mode, incremental currents from −300 to +500 pA were injected in 25 or 50 pA steps to explore the subthreshold and firing properties of the neurons. Off line analysis was performed using pClamp V9.2 (Molecular Devices, San Jose, CA, USA), Origin V7 (OriginLab, Northampton, MA, USA), Prism5 (GraphPad Software, SanD iego, CA, USA) and R (R Foundation for Statistical Computing, Austria). After recording, the slices were rinsed, fixed in paraformaldehyde (2%), and kept frozen at −20 • C until biocytin revelation.

Cell Reconstruction and Morphometric Analysis
After the recordings, sections were fixed with 4% paraformaldehyde, rinsed with 0.1 M phosphate-buffered saline, pH 7.4 (PBS) and incubated with streptavidin-conjugated Alexa Fluor ® 488 (Table S1) diluted 1:500 in PBS with 0.1% Triton X-100 and 0.02% sodium azide for 24 h in the dark. After washing with PBS, the sections were mounted on glass slides and coverslipped with an aqueous-based mounting medium.
Only completely Alexa Fluor ® 488 filled neurons were analyzed. Serial optical sections (0.1 µm step size) of each neuron were acquired with a laser-scanning confocal microscope (Leica, TCS NT) using a 63 × 1.3 numerical aperture oil-immersion. Images were acquired at high-resolution, with an image size of 1024 × 1024 pixels, and a voxel size of 0.27 × 0.27 × 1 µm for the x, y, and z axes, respectively. Image stacks were deconvolved to reduce signal blurring and each cell were three-dimensional reconstructed, traced, and analyzed using the image analysis system NeuronStudio (Icahn School of Medicine at Mount Sinai) [31]. For Sholl analysis, the number of intersections were counted between dendritic branches and a series of concentric spheres, starting at 10 µm from the soma and at radial increment of 1 µm. To quantify spine density, a digital zoom (6.7-fold augmentation) was applied and at least 10-15 dendritic segments (10 µm length) per branch order from each neuron were obtained. Each segment was manually inspected and appropriate corrections made using the NeuronStudio interface. Dendritic protrusions (thin, stubby, and mushroom spines) were classified according to their shapes: mushroom with large head and short neck; thin with thin head and long neck; and stubby with large head and no apparent neck.

Immunohistochemistry
Animals (n = 24) were transcardially perfused with 0.1 M PBS followed by 4% paraformaldehyde under deep sodium pentobarbital anesthesia (60 mg/kg, i.p.). The brains were rapidly removed, overnight post-fixed in the same fixative, cryoprotected in 30% sucrose in PBS (72 h) and frozen in dry ice. Rostro-caudal series of coronal sections (30 µm thick) were obtained with a freezing microtome (CM 1325, Leica, Weztlar, Germany) and stored in PBS containing 0.02% sodium azide. Free-floating sections were taken from the CPu (+1.0 to −0.30 mm from Bregma) and substantia nigra (−5.30 to −5.80 mm from Bregma) and processed for either single or double immunolabeling procedures.
Details regarding the antibodies used in the present study are listed in Table 1 (primary antibodies) and Table S1 (secondary antibodies). All primary antibodies were diluted in phosphate-buffered saline containing 0.2% Triton X-100 (PBS-TX) and 0.1% sodium azide.

Single Immunohistochemical Labeling
Free-floating brain sections were pre-treated for 15 min with 3% H 2 O 2 , rinsed in 0.1 M PBS and then incubated in the primary antibody (Table 1) for 24-48 h at room temperature. After being rinsed in PBS, the sections were incubated for 1 h in the appropriate secondary biotinylated antibody (Vector Laboratories, Burlingame, CA, USA) (Table S1)   incubated for 1 h at room temperature in a mixture of the appropriate secondary antibodies conjugated with Alexa 488 or Alexa 568 (Thermo Fisher, Waltham, MD, USA) (Table S1) yielding a green or red fluorescent signal, respectively. At the end of the staining, the sections were mounted on glass slides, coversliped with an aqueous-based mounting medium and observed with a Leica SP8 laser confocal microscopy (Leica, Wetzlar, Germany).

Microscopic Analysis and Semi-Quantification
Semi-quantitative analyses of optical density (OD) of immunoreactivity (IR) were performed, as described elsewhere [30], using the software ImageJ 1.48v (National Institutes of Health, NIH). The measures were performed form gray-scale photomicrographs obtained with a digital camera (DS-Fi1, Nikon, Tokyo, Japan) coupled to an optical microscope (Eclipse E400, Nikon, Tokyo, Japan) (40× objective). All OD values where corrected with the OD from an immunonegative area and data were expressed as mean percentage OD of control.

Western Blot
Animals (n = 16) were sacrificed by decapitation and the brains immediately removed. The substantia nigra was dissected and homogenized in a lysis buffer containing 2 mM orthovanadate and a mixture of protease inhibitors (Roche Diagnostics GmbH, Switzerland). After incubating on ice for 30 min, samples were centrifuged at 8000× g for 10 min at 4 • C. The supernatant was used for western blotting. Equal amounts of protein (50 µg) was loaded and resolved in 12 % SDS-polyacrylamide gel electrophoresis and then transferred onto PVDF membranes (Hybond-P, GE Healthcare, UK). Membranes were incubated overnight at 4 • C with a specific primary antibody (Table 1), followed by HRP-conjugated secondary antibody. Bands of the membranes were visualized using chemiluminescence detection (ECL, GE Healthcare, UK). Comparisons between the experimental groups were performed by determining bands immunoreactivity by densitometry with the image analyzing system ImageJ1.48v (NIH).

Statistical Analysis
Data are expressed as mean ± SEM. Statistical differences (P < 0.05) were assessed by one-way analysis of variance (ANOVA) followed by Bonferroni t test or Kruskal-Wallis analysis followed by Dunn's test for nonparametric data. Statistical analysis and graphs were generated using Prism 5 (GraphPad Software).

D 4 R Activation Results in MSNs Dendritic Arbors Stretching and Prevention of Morphine-Induced MSNs Dendritic Contraction
Morphine-induced structural plasticity has been previously demonstrated in several brain regions-e.g., NAc, prefrontal cortex or hippocampus [5,15]. To address whether morphine could also produce structural plasticity on striatal projection neurons of the CPu, we first analyzed MSNs dendritic arborization after 14 days of continuous drug treatment ( Figure 1A). We found that morphine apparently did not alter the total number of dendrites ( Figure 1B), neither the total dendrite length ( Figure 1C) nor the number of nodes ( Figure 1D). However, there was a pronounced and significant reduction in the dendritic volume (by 60%) ( Figure 1E). To further explore the impact of morphine on the dendritic arbor complexity, a morphological analysis was performed considering both the dendritic branch order and the distance from the soma. Morphine treatment produced both a remarkable shrinkage of proximal dendrites as a consequence of its shortening (>60% of reduction in the 1st, 2nd, and 3rd branch order) and volume diminishment (by 50% in the dendrites directly emanating from the soma) ( Figure 1C'-E').

D 4 R Activation Counteracts Dendritic Spine Depletion Induced by Morphine
We next analyzed the effects of the continuous treatment (14 days) with morphine and/or PD168,077 on MSNs spine density ( Figure 2A). Spine density was significantly reduced in the striatal MSNs of morphine-treated animals as compared to control rats (by 25%) ( Figure 2B). This spine loss was observed in distal dendrites (3rd, 4th, 5th, and 6th branch order) ( Figure 2C) and it was associated with the selective reduction in stubby (by 48%) and mushroom (by 42%) spines ( Figure 2D). The D 4 R agonist produced an even greater decrease in the overall spine density (by 45%) ( Figure 2B), but in this case it occurred throughout the entire dendritic arbor ( Figure 2C). The analysis of spine morphology indicated that such PD168,077-induced lower spine density was due to a decrease in the number of thin (by 40%), stubby (by 58%) and mushroom (by 60%) spines ( Figure 2D).
When PD168,077 was administered together with morphine, spine density was apparently reestablished (Figure 2A-C), as it was observed for thin-shaped spines ( Figure 2D). However, it should be noted that morphine + PD168,077 was unable to restore the loss of stubby spines and only partially did for mushroom spines ( Figure 2D).
Since spinophilin is a protein highly enriched in dendritic spines where it plays a key role in modulating spine density and synaptic activity [33], we evaluated its possible regulation by morphine and/or PD168,077. It was showed that the continuous treatment with morphine or PD168,077 alone downregulated striatal levels of spinophilin, which was specifically counteracted by the co-treatment of both drugs ( Figure 2E and Figure S1). Thus, the drugs-induced changes in spinophilin protein levels corroborated the modulation of spine density described above. Given the ability of D4R activation to prevent morphine-induced impairment of nigral dopaminergic signaling [16] and some of the molecular maladaptive changes related to drug addiction [25][26][27]32], we next studied whether the D4R agonist PD168,077

Changes in Striatal MSN Cell Bodies after Continuous Administration of Morphine and/or PD168,077
Given the specific remodeling of the MSNs dendritic arbors and dendritic spines in response to continuous treatment with morphine or/and PD168,077, we next examined whether these changes appear together with morphological alterations in the soma (Figure 3A). We observed that morphine or PD168,077 did not affect the morphology of the striatal MSNs somata, demonstrated by an absence of changes in the size (surface area and volume; Figure 3B,C) and shape (circularity; Figure 3D). However, the combined treatment with both drugs significantly increased the size of the striatal projecting neurons (surface area: by 33%; volume: by 55%) (Figure 3B,C).

Changes in Striatal MSN Cell Bodies after Continuous Administration of Morphine and/or PD168,077
Given the specific remodeling of the MSNs dendritic arbors and dendritic spines in response to continuous treatment with morphine or/and PD168,077, we next examined whether these changes appear together with morphological alterations in the soma ( Figure 3A). We observed that morphine or PD168,077 did not affect the morphology of the striatal MSNs somata, demonstrated by an absence of changes in the size (surface area and volume; Figure 3B,C) and shape (circularity; Figure 3D). However, the combined treatment with both drugs significantly increased the size of the striatal projecting neurons (surface area: by 33%; volume: by 55%) (Figure 3B,C).

D 4 R Activation during Continuous Treatment with Morphine Changes the Passive Properties and Excitability of Striatal MSNs
Since the continuous treatment with morphine and/or PD168,077 regulated dendritic arborization, dendritic spines density and soma size of striatal MSNs, we hypothesized that their electrical activity might be also altered. We used whole-cell patch-clamp recordings to evaluate the intrinsic properties and excitability of striatal MSNs from rats treated for 14 days with vehicle, morphine, PD168,077 or morphine + PD168,077 ( Figure 4A). The combined treatment with morphine + PD168,077 yielded significant differences in several passive and active MSNs membrane properties: (i) increased mem-brane capacitance ( Figure 4B); (ii) reduced membrane resistance ( Figure 4C); (iii) increased rheobase ( Figure 4F); and (iv) depolarized membrane potential ( Figure 4G). However, the action potential threshold (figfig:cells-1456217-f004D) and the first action potential latency ( Figure 4E) were not altered by any of the drug treatments.

D4R Activation During Continuous Treatment with Morphine Changes the Passive Properties and Excitability of Striatal MSNs
Since the continuous treatment with morphine and/or PD168,077 regulated dendritic arborization, dendritic spines density and soma size of striatal MSNs, we hypothesized that their electrical activity might be also altered. We used whole-cell patch-clamp recordings to evaluate the intrinsic properties and excitability of striatal MSNs from rats treated for 14 days with vehicle, morphine, PD168,077 or morphine + PD168,077 ( Figure 4A). The combined treatment with morphine + PD168,077 yielded significant differences in several passive and active MSNs membrane properties: i) increased membrane capacitance (Figure 4B); ii) reduced membrane resistance ( Figure 4C); iii) increased rheobase ( Figure 4F); and iv) depolarized membrane potential ( Figure 4G). However, the action potential threshold ( Figure 4D) and the first action potential latency ( Figure 4E) were not altered by any of the drug treatments. In addition to the changes in the passive properties, excitability was also reduced in the morphine + PD168,077 group compared with vehicle, morphine or PD168,077. The combined treatment group showed significantly smaller voltage-deflection ( Figure 4H) and slower firing frequency ( Figure 4I) in response to negative or positive current injection (50 pA steps), respectively.

Regulation of GIRK1 and GIRK2 Expression in the Striatal MSNs and Nigral Dopamine Neurons by Morphine and/or PD168,077 Treatments
Changes in the intrinsic and firing properties of striatal MSNs induced by the cotreatment with morphine + PD168,077 suggest a putative modulation of the heteromeric G protein-gated inward rectifier K + channels (GIRK), since these channels are the predominant channel open at the resting membrane potential [34] and it have been largely involved in the control of neuron excitability [35]. This hypothesis is supported by our previous results demonstrating a deregulation in the gene expression pattern of several GIRK subunits after the acute treatment with morphine and/or PD168,077 [36].  GIRK1 immunoreactivity (IR) was found in the striatal MSNs displaying a punctate labeling in the cytoplasm ( Figure 5A) and in most of the striatal interneurons (PV, SS, and ChAT) showing an intense reticulum-like distribution in the soma and primary dendrites ( Figure 5A,B). GIRK2 IR was no detected in the CPu ( Figure 5C). Both GIRK1 IR ( Figure 5D) and GIRK2 IR ( Figure 5E) were present in almost all dopaminergic cells of the SNc identified with an anti-TH antibody. GIRK1 IR showed a reticulum-like distribution in the cell bodies ( Figure 5D) whereas GIRK2 was demonstrated in both the perikarya and dendritic branches, even in those that extend down into the substantia nigra pars reticulata (SNr) ( Figure 5E).
The continuous treatment (7 days) with morphine or PD168,077 did not alter GIRK1 IR in the striatal MSNs. However, the combined administration of both drugs produced a significant increase in GIRK1 IR (by 23%) (Figure 5F). No changes in GIRK1 IR were detected in ChAT interneurons ( Figure S2A), which were identified by their morphological characteristics.
We also observed that the expression of GIRK1 and/or GIRK2 was regulated even after a single dose of morphine and/or PD168,007 in the CPu and SNc, although with a different pattern ( Figure S2B).

D 4 R Activation Counteracts Morphine-Induced Upregulation of MOR IR in the Striosome-Dendron Bouquets
A subset of striatal MSNs originates a direct striatonigral pathway that has been highlighted as crucial modulator of dopamine neuron function in the SNc [37][38][39]. These direct stritatonigral projections target dopaminergic neurons and its ventrally extending dendrites, where they form a specialized integrative unit that has been termed 'striosomedendron bouquet' [40]. High expression of MOR in the striosome-dendron bouquets [40] (and own observation) lead to the speculation that morphine and/or PD168,077 treatments could regulate the levels of this receptor as they do in the CPu [26]. The continuous treatment with morphine (7 days) induced a rise of MOR immunoreactivity (IR) (by 92%) that was completely prevented when morphine was administrated together with PD168,077 ( Figure 6A,B).
To obtain more information concerning both synaptic plasticity and transmission in the SN, we used western blot to measure the expression levels of synaptophysin, which is a presynaptic vesicle protein involved in the final steps of exocytosis and synapse formation [41]. Consistent with the previous results described above, it was observed that morphine upregulated synaptophysin (by 30%) in the substantia nigra. The administration of PD168,077 with morphine completely neutralized morphine-induced changes in the expression levels of this protein ( Figure 6C). drites, where they form a specialized integrative unit that has been termed 'striosomedendron bouquet' [40]. High expression of MOR in the striosome-dendron bouquets [40] (and own observation) lead to the speculation that morphine and/or PD168,077 treatments could regulate the levels of this receptor as they do in the CPu [26]. The continuous treatment with morphine (7 days) induced a rise of MOR immunoreactivity (IR) (by 92%) that was completely prevented when morphine was administrated together with PD168,077 ( Figure 6A,B). To obtain more information concerning both synaptic plasticity and transmission in the SN, we used western blot to measure the expression levels of synaptophysin, which is a presynaptic vesicle protein involved in the final steps of exocytosis and synapse formation [41]. Consistent with the previous results described above, it was observed that Kruskal-Wallis test followed by Dunn's method (B) or one-way ANOVA followed by Tukey test (C). Scale bar is 100 µm.

Discussion
The purpose of the present study was to determine whether the activation of D 4 R with the specific agonist PD168,077 could modulate morphine-induced neuroplasticity in the CPu. Prolonged exposure to morphine triggered structural plasticity in the dorsal striatal MSNs, which mainly consists of the retraction of the dendritic arbor and the loss of dendritic spines. It is also demonstrated that morphine upregulates MOR in the striatonigral inputs that exert a direct inhibition over nigral dopamine neurons-i.e., striosome-dendron bouquets. On the other hand, the continuous stimulation of D 4 R during morphine treatment also induced structural plasticity in the MSNs, in this case yielding an expansion of the dendritic arbor but restoring dendritic spine density. It was of special interest that the combined treatment of morphine with the D 4 R agonist PD168,077 altered the passive electrical properties and decreases the excitability of the striatal MSNs, along with changes in GIRK channels expression levels and the restoring of MOR expression in the striosomedendron bouquets.

Morphine-Induced Structural Plasticity in Striatal MSNs
There is strong evidence that morphine elicits structural and functional plasticity mostly in brain regions associated with reward, learning, or incentive motivation, such as the NAc, VTA, PFC, or hippocampus [5,15,42]. These neuroplastic changes persist for long after the discontinuation of morphine treatment. In fact, a recent research has found that morphine-induced alterations in proteins related to synaptic plasticity and cytoskeletal organization persist even six months after cessation of drug administration [43]. However, little was known about whether morphine also produced structural plasticity in the CPu, despite this region is critically involved in drug-related habits formation and consolidation of addiction [1,8,44].
Results of the present study clearly demonstrated that a continuous treatment with morphine triggers a retraction of the dendritic arbor and a significant loss of dendritic spines in the dorsal striatal MSNs. This observation is consistent with previous reports describing a similar pattern of long-term morphine-induced structural plasticity in several regions belonging to the reward circuit [2,5,12,45]. However, more recent studies have highlighted some contradictory data depicting an opposite effect, especially in the NAc, the orbital prefrontal cortex and the dentate gyrus of the hippocampus [13][14][15]46]. These differences could be attributed to several factors, such as the paradigm of drug administration, the animal sacrifice time point, the brain region analyzed, or even the cell type studied.
Interestingly, we observed that the retraction of dendrites specifically occurred by a decrease in the size (length and volume) of the proximal dendrites, while the spines loss was attributable to a selective downregulation of mushroom and stubby spines on distal dendrites. The concurrence of these two forms of structural plasticity could have a major impact on MSNs activity. Firstly, because signals propagation effectiveness throughout dendrites is highly dependent on their morphology [47][48][49][50]. Secondly, because dendritic spines remodeling is tightly linked to synaptic strength, and therefore to the synaptic plasticity phenomena [51]. Thus, the selective loss of mushroom spines after prolonged exposure to morphine could reflects the induction of long-term depression (LTD) processes, as previously demonstrated in excitatory inputs to the dorsal striatum [52,53]. It should be noted that these two structural changes took place in specific subregions of the dendritic arbor, revealing a dichotomy between proximal and distal dendrites. Regarding this, the existence of a topographic organization of the striatal MSNs inputs has been largely suggested although not fully unraveled. Proximal dendrites seem to mainly receive intrinsic shaft connections from striatal interneurons and collateral projections from MSNs, whereas distal dendrites preferentially support extrinsic projections from the cortex, thalamus, or midbrain [54,55]. In this perspective, it is likely that morphine, by promoting morphological changes and therefore affecting synaptic inputs, has a significant impact on the function of the MSNs of the dorsal striatum and ultimately on their output.
Our results show that, although morphine can induce an overall structural plasticity in the MSNs of the CPu, it is unable to modify their passive membrane properties and intrinsic excitability. Interestingly, previous studies have demonstrated that repeated morphine exposure specifically alters the activity of the NAc MSNs which expressed the dopamine D2 receptors (D 2 R) [56,57]. Thus, as in our analysis we did not identify striatal MSNs subpopulations-e.g., direct-and indirect-pathway projection neurons-it remains to be investigated in future works whether morphine specifically mediates changes in one of them. On special interest will also be to reveal the specific action of morphine on the striatonigral projection neurons located in the striosomal compartment since this neural population express MOR and provide direct inputs on nigral dopamine neurons [37][38][39][40]58] and we have observed a morphine-induced increase in MOR expression in the striosomedendron bouquets.

D 4 R Activation Alter Dendritic Arbor Complexity and Spine Density of the Striatal MSNs
Dopamine action on striatal MSNs through its interaction with dopamine receptors has a major role in both structural and synaptic plasticity occurring in these neurons [59]. In fact, the existence of opposite processes controlling synaptic plasticity induced by dopamine in form of LTP or LTD are postulated to be directed by D 1 R or D 2 R in the direct-and indirect-pathway MSNs, respectively [60][61][62]. These data support the classical model for the independence of the two main striatal pathways outputs [63]. However, this model is currently under review since recent data are quite conflictive. As an example, a recent report has described that D 1 R is essential for the maintenance of spine plasticity in the direct-pathway MSNs but it also affects indirect-pathway neurons [64]. This could be explained on the base of intrinsic connectivity between direct-and indirect-pathways MSNs or the cross-talk of these neurons via interneurons [62]. In addition, several works confirm that a certain proportion of MSNs co-express D 1 R and D 2 R [65,66].
The role of other receptor subtypes different to D 1 R and D 2 R has been largely undervalued in the context of the CPu despite their expression in both direct-and indirect-pathways MSNs. Thus, the existence of D 4 R and D 5 R in the CPu [28,67] opens up the possibility to discover new dopamine signaling mechanisms underlying striatal plasticity. According to our results, we suggest a new role for the D 4 R in mediating the regulation of MSNs morphology by dopamine. Here, we show that prolonged D 4 R stimulation by itself expands the MSNs dendritic arbors, as the activation of these receptors increase both the number of distal dendritic branches and also of their length. This dendritic reshaping could be related to the ability of D 4 R to modulate the expression of the neurotrophic factor BDNF [36] and the transcription factors CREB and ∆FosB [25], which, in turn, regulates the transcription of numerous genes encoding for cytoskeleton regulatory proteins [5]. Paradoxically, D 4 R stimulation also induces a downregulation of all types of dendritic spines (i.e., thin, stubby and mushroom), which occurs across the entire dendritic arbor. However, despite the morphological alterations, the passive membrane properties and excitability remain unchanged. This effect differs from that previously described for D 2 R, since overexpression of this receptor increases the excitability of the MSNs and decreases the complexity and length of their dendritic arbors [68].

D 4 R Activation Modulates Morphine-Induced Plasticity
D 4 R is at present under the focus of interest because of its possible role in cocaine, amphetamine, nicotine, and alcohol addiction [23,24,[69][70][71]. It also exists strong evidence that stimulation of the D 4 R with the selective agonist PD168,077 disrupts morphine addiction without affecting its analgesic properties [16]. This effect seems to be the result of an antagonistic receptor-receptor interaction involving a hypothetical D 4 R-MOR heterodimer [16,26], which could exist in several regions were both receptors are co-expressed-striosomal compartment of the CPu and the SNr [28,72].
The current morphological results demonstrated that D 4 R stimulation reversed morphine-induced alterations in thin and mushroom spines, but not in stubby spines. The combined D 4 R agonist and morphine treatment also leads to an expansion of the dendritic arbor that is similar to that observed after the exclusive treatment with the D 4 R agonist. In agreement with these results, transcriptome analysis following acute administration of morphine and PD168,077 provided indications for the modulation of several genes related with dendritic spine dynamic and synaptic formation-Homer1, Pfn2 and Slitrk1 [73][74][75] that could be also regulated after their chronic administration. Taken together, these results suggest that a cross-talk between D 4 R and MOR could occur to modulate specifically structural plasticity of dendritic spines.
A parallel decrease in MSNs excitability was also observed upon morphine and PD168,077 co-treatments, which was accompanied by related membrane property changes. According with the increase in membrane capacitance, our results showed an enlargement of the MSNs cell surface. On the other hand, changes in the intrinsic excitability could be related with an alteration in the inwardly rectifying K + currents through a modified function of GIRK channels [57] upon MOR and/or D 4 R activation [76,77]. As a proof of concept, we checked changes in the protein expression levels of two of the GIRK channels subunits-GIRK1 and GIRK2-which were previously highlighted by its presumed role in the D 4 R modulation over morphine effects [36]. However, only a significant upregulation of GIRK1 was observed in the CPu after morphine and PD168,077 treatments as no expression change of GIRK2 was found. It is well established that GIRK1 requires the interaction with other GIRK subunits to reach the cell surface and therefore to form functional channels [35,78,79]. Thus, further studies will be needed to clearly characterize the subunits that form the GIRK channel heterotetramers that operates in the dorsal striatal MSNs and its function regarding the regulation of morphine effects.
In addition to the modulatory effect of D 4 R over morphine-induced plasticity in the CPu, we here also observe two indications of neuroadaptative changes in both the nigral dopamine cells and the striatal projections which regulate its function. Firstly, the co-administration of morphine and the D 4 R agonist PD168,077 restores the expression levels of MOR in the GABAergic terminals coming up from the striosomal MSNs which form a network surrounding dopamine nerve cells and its ventrally extending dendrites-'striosome-dendron bouquet' [40]. Regarding this, our own previous observation also demonstrated a D 4 R modulation of MOR in the striosomal compartment [26]. Secondly, D 4 R activation counteracts morphine-induced downregulation of GIRK1 and GIRK2 subunits in nigral dopamine cells, which could be associated, as occurs in the hypothalamus, to a reduction in the GIRK channel function [77]. Our hypothesis, which clearly requires further evaluations, is that D 4 R prevents morphine-induced MOR sensitization in the striatal MSNs, leading to a restoring of GABA signaling [27] from the striatum to the SNc which ultimately could reestablish nigral dopamine cell excitability and a normal dopamine output toward the CPu [16,42].
In conclusion, in combination with our previous works that evidence that the stimulation of the D 4 R prevents morphine-induced reward but not analgesia, we have demonstrated that D 4 R appears as a key regulator of morphine-induced plasticity in the CPu. Therefore, it is plausible that D 4 R may disrupts long-term effects of morphine that drive ventral-to-dorsal striatal shifts in addiction consolidation. Thus, pharmacological strategies involving D 4 R activation might take advantages as a potential therapeutic strategy for the safety prescription of morphine for pain relief.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10.3 390/cells11010031/s1, Figure S1: Changes in spinophilin IR in the CPu after continuous administration (14 days) of morphine and/or PD168,077. Figure S2: Changes in GIRK1 IR in the striatal ChAT interneurons and changes in GIRK1 and GIRK2 IR in the CPu and SNc after acute administration of morphine and/or PD168,077. Table S1: Secondary antibodies used for immunohistochemistry and western blot.

Informed Consent Statement: Not applicable.
Data Availability Statement: The datasets in this study are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.