MPTP Impairs Dopamine D1 Receptor-Mediated Survival of Newborn Neurons in Ventral Hippocampus to Cause Depressive-Like Behaviors in Adult Mice

Parkinson’s disease (PD) is characterized by motor symptoms with depression. We evaluated the influence of dopaminergic depletion on hippocampal neurogenesis process to explore mechanisms of depression production. Five consecutive days of 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) injection in mice (MPTP-mice) reduced dopaminergic fibers in hippocampal dentate gyrus (DG). MPTP-mice exhibited depressive-like behaviors later for 2–3 weeks. BrdU was injected 4 h after last-injection of MPTP. BrdU-positive (BrdU+) cells in dorsal (d-DG) and ventral (v-DG) DG were examined on day 1 (D1), 7 (D7), 14 (D14) and 21 (D21) after BrdU injection. Fewer D7-, D14- and D21-BrdU+ cells or BrdU+/NeuN+ cells, but not D1-BrdU+ cells, were found in v-DG of MPTP-mice than in controls. However, the number of BrdU+ cells in d-DG did not differ between the both. Loss of doublecortin-positive (DCX+) cells was observed in v-DG of MPTP-mice. Protein kinase A (PKA) and Ca2+/cAMP-response element binding protein (CREB) phosphorylation were reduced in v-DG of MPTP-mice, which were reversed by D1-like receptor (D1R) agonist SKF38393, but not D2R agonist quinpirole. The treatment of MPTP-mice with SKF38393 on days 2–7 after BrdU-injection reduced the loss of D7- and D21-BrdU+ cells in v-DG and improved the depressive-like behaviors; these changes were sensitive to PKA inhibitor H89. Moreover, the v-DG injection of SKF38393 in MPTP-mice could reduce the loss of D21-BrdU+ cells and relieve the depressive-like behaviors. In control mice, the blockade of D1R by SCH23390 caused the reduction of D21-BrdU+ cells in v-DG and the depressive-like behaviors. Our results indicate that MPTP-reduced dopaminergic depletion impairs the D1R-mediated early survival of newborn neurons in v-DG, producing depressive-like behaviors.


INTRODUCTION
Parkinson's disease (PD) is a neurodegenerative disorder characterized by motor symptoms and a progressive loss of dopaminergic neurons (Rodriguez-Oroz et al., 2009). Major depression is present in approximately 30-50% of PD patients (Bower et al., 2010;Nègre-Pagès et al., 2010). However, the underlying mechanisms remain unclear.
The hippocampal volume is smaller in patients with depression than in normal subjects of the same-age (Videbech and Ravnkilde, 2015). Electroconvulsive therapy in refractory depression can increase hippocampal volume. Growing evidence suggests that neurogenesis continues throughout adulthood within the hippocampal dentate gyrus (DG; Toni et al., 2008). The newly generated neurons can integrate into hippocampal circuitry, thereby maintaining a functional structure, which is required for mood control and antidepressant efficacy (Petrik et al., 2012). The deficits in adult neurogenesis in the postmortem brains of PD patients have recently attracted much attention. For example, Höglinger et al. (2004) have reported the deficits in the hippocampal neurogenesis of PD patients. The hippocampal cell proliferation capacity is decreased in PD patients (Guiard et al., 2009). A reduction in neuronal precursor cells has been observed in the subgranular zone (SGZ) of patients with PD (Borta and Höglinger, 2007).
The hippocampus and the midbrain dopaminergic neurons of the ventral tegmental area (VTA) form a functional loop (Lisman and Grace, 2005). Dopaminergic fibers, originating mainly from VTA (Gasbarri et al., 1997), directly contact the newborn cells in DG, but are sparse in the granule cell layer or adjacent hilus (Höglinger et al., 2004). Neuronal precursor cells in the hippocampal DG of adult mammals express all dopaminergic receptors (DRs), which receive dopaminergic afferents (Takamura et al., 2014). Dopamine (DA) has been reported to play an important role in the regulation of endogenous neurogenesis in the adult mammalian brain. The activation of D1-like receptors (D1R) promotes the survival of newborn cells in the adult hippocampus (Takamura et al., 2014). Winner et al. (2009) have observed that a D2-like receptor (D2R) agonist in DA-depleted rats increases the proliferation of neural precursor cells in the subventricular zone (SVZ) but not in SGZ. A large body of evidence has established that the DA depletion in rodents decreases the cell proliferation and survival of neuronal precursor cells in DG (Höglinger et al., 2004;Khaindrava et al., 2011). However, other data have revealed that the DA depletion increases or has no effect on cell proliferation (Oizumi et al., 2008;Park and Enikolopov, 2010).
In rodents and primates, the changes in afferent and efferent connectivity along the longitudinal axis of the hippocampus suggest distinct functions of the dorsal and ventral hippocampus. The dorsal DG (d-DG) receives projections arising in the lateral and caudomedial portion of entorhinal cortex, whereas the ventral DG (v-DG) receives inputs from the rostromedial region of the entorhinal cortex (Dolorfo and Amaral, 1998). The innervation density of dopaminergic plexuses is very high in the ventral hippocampus, but is low in the dorsal portion (Bjarkam et al., 2003). 1-Methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP) is commonly used to prepare an animal model of PD as it selectively induces cell death of dopaminergic neurons (Schober, 2004). The MPTP lesion can be used for behavioral studies of affective disorders in C57BL/6 mice (Gorton et al., 2010). To investigate whether the DA depletion in PD brain impairs the hippocampal neurogenesis to cause the depression, we in this study examined the affective behaviors and the neurogenesis process of d-DG and v-DG (including the proliferation of stem cells, the differentiation of progenitor cells and the survival of newborn neurons) in mice treated with MPTP (MPTP-mice), and explored the correlation between hippocampal neurogenesis deficits and depressive-like behaviors. The present study provides evidence that MPTP-reduced dopaminergic afferents impairs the D1Rmediated early survival of newly generated neurons in v-DG, which may be responsible for the production of depressive-like behaviors.

Mice
The use of animals was approved by Institutional Animal Care and Use Committee of Nanjing Medical University and was performed in accordance with the experimental animal guidelines of Laboratory Animal Research Institute. Eight-weekold male C57BL/6 mice (24-26 g; Oriental Bio Service Inc., Nanjing, China) were used at the beginning of the experiment. The mice were maintained under constant environmental conditions (temperature 23 ± 2 • C, humidity 55 ± 5% and 12:12 h light/dark cycle) in Animal Research Center of Nanjing Medical University with free access to food and water.

Behavioral Examination
A single cohort of animals was used for the following test sequence: open-field test (OFT) → Forced swim test (FST) → Tail suspension test (TST; Zhou et al., 2014). All behavioral data were captured by a video-monitor and analyzed using TopScan Lite 2.0 (Clever Sys., Reston, VA, USA).
FST was performed in a glass cylinder (300 mm high, 280 mm in diameter) that was filled with water (25 ± 1 • C) to a height of 20 cm. Total immobility time during a 6 min test was scored. Mice were considered to be immobile when they stopped struggling and moved only to remain floating in the water, while keeping their heads above water.
TST was performed by using adhesive tape to attach the tail to a rod that was 60 cm above the floor. Trials were conducted for a period of 6 min, during which the immobility time was recorded.
Morris water maze task (MWM) was performed in a blackcolored plastic pool (diameter = 120 cm) at 20 ± 1 • C. A cylindrical platform (diameter = 7 cm) was placed 0.5 cm below the surface of water. Each mouse was randomly released from four different quadrants and allowed to swim for 90 s. Four trials were conducted each day with an intertrial interval of 30 min. Average swimming speed (m/s) and latency (s) to reach the platform were scored for all trials. If a mouse could not reach the platform within 90 s, the experimenter gently assisted the mouse onto the platform and allowed it to remain there for 15 s.

Western Blot Analysis
After the dorsal and ventral hippocampal slices were harvested, the DG regions were micro-dissected and stored at −80 • C until assayed. Protein was extracted and the concentration of protein was determined using a bicinchoninic acid (BCA) protein assay kit (Pierce, IL, USA). Protein (40 µg) was separated by 10% acrylamide denaturing gels (SDS-PAGE) and transferred to membranes. The membranes were incubated with rabbit anti-protein kinase A (PKA) phosphorylation (1:1000, Millipore) and rabbit anti-cyclic AMP-response element binding protein (CREB) phosphorylation (1:1000, Santa Cruz, CA, USA). Then, the membranes were incubated with horseradish peroxidase-labeled goat anti-rabbit antibody (1:5000, Santa Cruz, CA, USA), and developed using an enhanced chemiluminescence detection kit (Millipore). After visualization, the blots were stripped by incubation in stripping buffer for 15 min, and then incubated with rabbit anti-PKA antibody (1:1000, Millipore) and rabbit anti-CREB antibody (1:1000, Santa Cruz, CA, USA). An internal control was performed using mouse anti-β-actin antibody (1:2000, Cell Signaling, Danvers, MA, USA). Western blot bands were scanned and analyzed with the ImageJ analysis software package (NIH).

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Total RNA of v-DG and d-DG was isolated using TRIzol reagent (Invitrogen, Camarillo, CA, USA) and reverse-transcribed into cDNA using a PrimeScript RT reagent kit (Takara, China) for quantitative PCR (ABI Step One Plus, Foster City, CA, USA) in the presence of a fluorescent dye (SYBR Green I; Takara). The relative expression of genes was determined using the 2 − ct method with normalization to the GAPDH expression. The primer sequences of D1R and D2R mRNA were designed as described in a previous publication (Kim et al., 2010).

Statistical Analysis
The group data were expressed as the means ± standard error (SEM). All statistical analyses were performed using SPSS software, version 16.0 (SPSS Inc., Chicago, IL, USA). Differences among means were analyzed using the Student's t-tests or one/two-factor analysis of variance (ANOVA) or repeatedmeasures ANOVA, followed by a post hoc Bonferroni test. Differences of p < 0.05 were considered statistically significant.
The spatial memory was further examined using the place learning of the MWM on days 22-26 post-MPTP. Repeatedmeasures ANOVA revealed that the escape latency of hidden platform progressively decreased with training days in all groups (F (4,72) = 59.308, p < 0.001; Figure 1F), which was not affected by the MPTP-injection (F (1,18) = 0.049, p = 0.828).
Doublecortin (DCX), a microtubule-associated protein, is specifically expressed in immature neuroblasts and progenitor cells (Francis et al., 1999). To evaluate the survival of young newborn neurons, we examined the number of DCX + cells in the hippocampal DG on day 7 post-MPTP. The results showed that the number of DCX + cells was significantly reduced in v-DG (p < 0.01, n = 10; Figure 2F), but not in d-DG (p > 0.05, n = 10), compared with those in control mice.
To examine the differentiation of precursor cells, we further examined the number and relative proportions of D21-BrdU + cells expressing NeuN, a mature neuron marker, or glial fibrillary acidic protein (GFAP), a glial marker. The number of D21-BrdU + /NeuN + cells in v-DG of MPTP-mice was less than that in control mice (p < 0.01, n = 10; Figure 2G), whereas there was no difference in d-DG between the both (p > 0.05, n = 10). In contrast, the number of D21-BrdU + /GFAP + cells in v-DG (p > 0.05, n = 10; Figure 2H) or d-DG (p > 0.05, n = 10) did not differ between MPTP-mice and control mice.
The administration of D1R antagonist SCH23390 for 2 days to control mice decreased the phosphorylation of PKA (p < 0.01) and CREB (p < 0.01, n = 8) in v-DG. Consistently with the results of a report by Zhao et al. (2013), the activation of D2R by quinpirole led to a decline in the phosphorylation of PKA (p < 0.05) and CREB (p < 0.05, n = 8) in v-DG of control mice, whereas the D2R antagonist L-sulpiride had no effects (p > 0.05, n = 8). The treatment of MPTP-mice with quinpirole showed a tendency to decrease the phosphorylation of PKA and CREB in v-DG, but the difference had no statistical significance (p > 0.05, n = 8). In contrast, MPTP-mice did not show the changes in the levels of the PKA and CREB phosphorylation (p > 0.05, n = 8) in d-DG compared with those of controls, which was not affected by SKF38393 (p > 0.05, n = 8) or quinpirole (p > 0.05, n = 8). Furthermore, the phosphorylation of PKA or CREB in d-DG of control mice was not altered by the application of SCH23390 (p > 0.05, n = 8) or L-sulpiride (p > 0.05, n = 8).

MPTP Reduces the D1R-Mediated Survival of Newborn Neurons
The newborn cells in the adult hippocampal DG express functional D1R and D2R (Takamura et al., 2014). To test whether the down-regulation of D1R-activated PKA-CREB signaling affects the survival of young newborn neurons, MPTP- mice were treated with SKF38393 or quinpirole daily on days 2-7 after the BrdU-injection, and control mice received the treatment with SCH23390, L-sulpiride or quinpirole. Then, we examined the number of D7-and D21-BrdU + cells, respectively ( Figure 4A). As shown in Figures 4B,C, the treatment with SKF38393 in MPTP-mice not only rescued the loss of D7-BrdU + cells (p < 0.01, n = 10), but also corrected the reduction of D21-BrdU + cells (p < 0.01, n = 10) in v-DG, which was sensitive to H89 (p < 0.05, n = 10). However, the administration of quinpirole to MPTP-mice had no effects on the reduced D7-or D21-BrdU + cells in v-DG (p > 0.05, n = 10). Interestingly, the treatment of control mice with either SCH23390 (p < 0.01, n = 10) or quinpirole (p < 0.05, n = 10), but not L-sulpiride (p > 0.05, n = 10), caused a decrease in the number of D7-BrdU + cells in v-DG, which was associated with the reduction of D21-BrdU + cells (SCH23390: p < 0.01, n = 10; quinpirole: p < 0.05, n = 10). In contrast, the number of D7-or D21-BrdU + cells in d-DG of control mice was not affected by SCH23390, quinpirole or L-sulpiride (p > 0.05, n = 10). However, the application of SKF38393 on days 16-21 after the BrdU-injection had no effect on the reduced D21-BrdU + cells in v-DG of MPTP-mice (p > 0.05, n = 10; Figure 4D). At the same time, the treatment of control mice with SCH23390 did not change the number of D21-BrdU + cells in v-DG (p > 0.05, n = 10). Additionally, the administration of SKF38393 for 6 days post-MPTP could increase the number of DCX + cells in v-DG (p < 0.01, n = 10; Figure 4E), while the activation of D2R by quinpirole could not (p > 0.05, n = 10). Similarly, the treatment of control mice with SCH23390 (p < 0.01, n = 10), but not L-sulpiride (p > 0.05, n = 10), caused a reduction in the number of DCX + cells in v-DG.

Association of MPTP-Impaired Neurogenesis With Depressive-Like Behaviors
To explore whether the impaired neurogenesis in MPTP-mice is related to their depressive-like behaviors, MPTP-mice were treated with SKF38393 or quinpirole for 18 days starting from the first-injection of MPTP, and control mice received SCH23390 and quinpirole for 18 days ( Figure 5A). To avoid the direct effects of drugs, the affective behavioral tests were performed on days 6-7 after the end of the drugs administration. The results showed that the treatment of MPTP-mice with SKF38393, but not quinpirole, corrected the prolongation of the immobility time in FST (p < 0.05, n = 10; Figure 5B) and TST (p < 0.01, n = 10; Figure 5C), which was blocked by H89 (FST: p < 0.05; TST: p < 0.05, n = 10). In addition, the administration of SCH23390 (FST: p < 0.01; TST: p < 0.01, n = 10) or quinpirole (FST: p < 0.05; TST: p < 0.05, n = 10) to control mice caused an increase in the immobility time of FST and TST.

DISCUSSION
The present study provides evidence that the MPTP-induced decline of dopaminergic afferents in v-DG impairs the D1Rmediated early survival of newly generated neurons, and this deficit in hippocampal neurogenesis is associated with depressive-like behaviors.
The adult DG contains at least two types of proliferating immature cells: GFAP + radial glia-like cell (B-cell) and GFAP − cell (C-cells). The B-cells, as stem cells, generate frequently dividing transit-amplifying C-cells. Asymmetrically divided C-cells are more likely to be progenitor cells and can differentiate into neural precursors or DCX + neuroblasts (A-cells; Zhao et al., 2008). D2R is predominantly expressed in C-cells, while A-cells express both D1R and D2R (Höglinger et al., 2004). The activation of D2R has been reported to stimulate C-cell proliferation (Yang et al., 2008). The number of D1-BrdU + cells in v-DG or d-DG of MPTP-mice was not altered, although the dopaminergic neurons in VTA were reduced by approximately 25% on day 1 post-MPTP. The microtubuleassociated protein DCX expression is specific to newly generated neurons, but not to glial cells and undifferentiated cells, which reaches a peak during the second week and is downregulated concomitantly with the appearance of NeuN (Rao and Shetty, 2004). DCX, as an immature neuron marker, is used to evaluate the survival of young newborn neurons (Kim et al., 2009). MPTP-mice had a significant decrease in the DCX + cells in v-DG, which could be rescued by the activation of D1R rather than D2R. In addition, the blockade of D1R caused a decrease of the DCX + cells in v-DG of control mice. The number of D7-BrdU + cells in v-DG of MPTP-mice was reduced. Although fewer D14-or D21-BrdU + cells were found in v-DG of MPTP-mice than those in control mice, the percentage of reduced BrdU + cells (D7 → −25% D14 → −57% D21) in MPTP-mice did not differ from control mice (D7 → −20% D14; D14→ −52% D21). Therefore, it is likely that the MPTP-induced DA depletion mainly impairs the early survival of newborn neurons in v-DG. Khaindrava et al. (2011) have reported that DA depletion causes deficits in the survival of newborn cells in DG. Our pharmacological experiments support this idea by showing that in MPTP-mice, the D1R agonist when administered on days 2-7 after BrdU-injection rescued the loss of D7-or D21-BrdU + cells, but when administered on days 16-21 after BrdU-injection, it failed to alter the reduced D21-BrdU + cells in MPTP-mice. Moreover, the blockade of D1R on days 2-7 after BrdU-injection reduced the number of D7-and D21-BrdU + cells in control mice. The absolute number of D21-BrdU + /NeuN + cells and proportion of these  cells relative to the total D21-BrdU + cells were reduced in v-DG of MPTP-mice. The number of BrdU + /GFAP + cells in v-DG of MPTP-mice did not differ from control mice, and the ratio of BrdU + /GFAP + cells was higher than the ratio in control mice. Thus, it is conceivable that the MPTP-induced DA depletion does not affect the differentiation of precursor cells. On the other hand, the number of BrdU + cells, DCX + cells, BrdU + /NeuN + cells or BrdU + /GFAP + cells in d-DG did not differ between control and MPTP-mice, which was not affected by the activation or blockade of D1R and D2R. Consistently with the report by Bjarkam et al. (2003), there was few dopaminergic projections in d-DG, indicating that the neurogenesis process in d-DG may be insensitive to the dopaminergic regulation. The administration of D1R agonist to MPTP-mice could rescue the loss of D7-BrdU + cells in a PKA-dependent manner, but the D2R agonist did not. The major signaling cascaded by D1R activation is the cAMP-PKA pathway in the hippocampus (Arnsten and Dudley, 2005). PKA-CREB signaling is one of the most important pathways for promoting neuron regeneration in DG (Miyamoto et al., 2009). The activation of PKA induces the phosphorylation of CREB at Ser-133, thereby facilitating the gene transcription of key molecules such as c-Fos, Jun-B, Bcl-2, GDNF and neurotrophins to regulate neuronal survival and regeneration. Inhibition of phosphodiesterase-4, an enzyme that catalyzes the hydrolysis of cAMP, stimulates the activation of CREB and increases the survival of newborn cells (Nakagawa et al., 2002). In this study, we observed the decline of PKA and CREB phosphorylation in v-DG of MPTP-mice without changes in the expression of D1R. In v-DG of MPTP-mice, the decrease of PKA and CREB phosphorylation could be corrected by the D1R agonist and the protection of D1R agonist on the D7-BrdU + cells was blocked by the inhibition of PKA. The blockade of D1R caused the loss of D7-BrdU + cells and decline of PKA and CREB phosphorylation in v-DG of control mice. The D2R antagonist haloperidol has been shown to decrease (Wakade et al., 2002), increase (Kippin et al., 2005) or not affect (Malberg et al., 2000) hippocampal neurogenesis. Our results showed that the activation of D2R reduced the PKA and CREB phosphorylation and the D7-BrdU + cells in v-DG of control mice. Therefore, it is proposed that the MPTP-induced DA depletion through the down-regulation of D1R-induced PKA-CREB signaling impairs the early survival of newborn neurons in v-DG.
It is widely accepted that the hippocampal neurogenesis is required for mood control (Petrik et al., 2012) and cognitive performance (Rola et al., 2004). Selectively impairing the adult neurogenesis by telomerase inhibitor could cause the depressionlike behaviors in mice (Zhou et al., 2011a). Irradiation of young animals impairs hippocampal neurogenesis that is associated with cognitive deficits (Rola et al., 2004). The newly generated neurons are integrated into the neuronal circuitry within 3-4 weeks after birth (Zhao et al., 2006) to regulate hippocampal output (Ming and Song, 2011). Importantly, we have observed the depressive-like behaviors in MPTP-mice for 2-3 weeks later after DA depletion. The timing of depressive-like behaviors in MPTP-mice seems to be coincident with the loss of mature newborn neurons. Interestingly, MPTP-mice did not appear affected by the spatial cognitive deficits, although they showed depressive-like behaviors. One possible explanation may be a difference in the regions of impaired neurogenesis in hippocampal DG. The idea may be supported by the facts that the lesions in the dorsal hippocampus affect spatial learning and memory (Moser et al., 1995), while lesions of the ventral hippocampus lead to anxiety and depressive-like behaviors (McHugh et al., 2004). The antidepressants may exert their behavioral effects by increasing neurogenesis in v-DG (Banasr et al., 2006). In particular, the v-DG injection of SKF38393 in MPTP-mice not only reduced the loss of D21-BrdU + cells but also relieved the depressive-like behaviors. In control mice, the v-DG injection of either SCH23390 or H89 caused the reduction of D21-BrdU + cells in v-DG, which was companied by the production of depressive-like behaviors. These results indicate a possible cause-and-effect relationship between the MPTP-impaired neurogenesis and the MPTP-induced affective disorder. In addition, the ventral hippocampus is involved in regulation of the hypothalamic-pituitary-adrenal axis (Herman et al., 1995), buffering the stress response (Snyder et al., 2011). The basolateral amygdaloid complex receives the hippocampal afferents (Supcun et al., 2012). Hippocampal dysfunction affects the synaptic plasticity of the basolateral amygdaloid complex, which is important for the acquisition and consolidation of fear memories (Goosens and Maren, 2002) and the extinction of learned fear (Dalton et al., 2012).
Forebrain DA circuitry has been studied by two largely independent researchers: a nigrostriatal DA system that originates in the substantia nigra (SN) and a mesolimbic DA system that originate in the VTA. MPTP selectively destroys the dopaminergic neurons in the SN and VTA (Lu et al., 2014;Hong et al., 2015; the present study). Target neurons in the primary terminal field of the SN are striatal medium spiny neurons. It is well known that the MPTP-induced loss of dopaminergic neurons in SN pars compacta through the striatal DA depletion impairs the motor functions including slowness, rigidity, resting tremor and postural instability (Fahn, 2003). The projection fields of VTA DA neurons are the hippocampus or medial prefrontal cortex, which contributes to the emotional regulation (Russo and Nestler, 2013). Tye et al. (2013) have reported that the selective inhibition of VTA DA neurons acutely induces multiple distinct depression-like behaviors. Furthermore, the dopaminergic fibers from VTA dopaminergic neurons have been reported to contact directly with the newborn cells in DG (Höglinger et al., 2004). Taken together, the results in the present study give an indication that the MPTP-impaired VTA dopaminergic neurons leading to the dopaminergic deficiency in v-DG, impairs the D1R-mediated early survival of newborn neurons.
The present study provides evidence that the MPTP-induced dopaminergic depletion impairs the D1R-mediated early survival of newly generated neurons in v-DG; this impairment of hippocampal neurogenesis is associated with the production of the depressive-like behaviors. Thus, D1R agonists may be candidate substrates for treating PD depression.

AUTHOR CONTRIBUTIONS
In this study, TZ performed the immuno-staining, western blotting and all statistical analysis. JH finished the behavioral examination and RT-PCR analysis. TD carried out the animal care. LC designed the experiment and finished the manuscript.