Medial prefrontal cortical PPM1F alters depression‐related behaviors by modifying p300 activity via the AMPK signaling pathway

Abstract Aims Protein phosphatase Mg2+/Mn2+‐dependent 1F (PPM1F) is a serine/threonine phosphatase, and its dysfunction in depression in the hippocampal dentate gyrus has been previously identified. Nevertheless, its role in depression of another critical emotion‐controlling brain region, the medial prefrontal cortex (mPFC), remains unclear. We explored the functional relevance of PPM1F in the pathogenesis of depression. Methods The gene expression levels and colocalization of PPM1F in the mPFC of depressed mice were measured by real‐time PCR, western blot and immunohistochemistry. An adeno‐associated virus strategy was applied to determine the impact of knockdown or overexpression of PPM1F in the excitatory neurons on depression‐related behaviors under basal and stress conditions in both male and female mice. The neuronal excitability, expression of p300 and AMPK phosphorylation levels in the mPFC after knockdown of PPM1F were measured by electrophysiological recordings, real‐time PCR and western blot. The depression‐related behavior induced by PPM1F knockdown after AMPKα2 knockout or the antidepressant activity of PPM1F overexpression after inhibiting acetylation activity of p300 was evaluated. Results Our results indicate that the expression levels of PPM1F were largely decreased in the mPFC of mice exposed to chronic unpredictable stress (CUS). Behavioral alterations relevant to depression emerged with short hairpin RNA (shRNA)‐mediated genetic knockdown of PPM1F in the mPFC, while overexpression of PPM1F produced antidepressant activity and ameliorated behavioral responses to stress in CUS‐exposed mice. Molecularly, PPM1F knockdown decreased the excitability of pyramidal neurons in the mPFC, and restoring this low excitability decreased the depression‐related behaviors induced by PPM1F knockdown. PPM1F knockdown reduced the expression of CREB‐binding protein (CBP)/E1A‐associated protein (p300), a histone acetyltransferase (HAT), and induced hyperphosphorylation of AMPK, resulting in microglial activation and upregulation of proinflammatory cytokines. Conditional knockout of AMPK revealed an antidepressant phenotype, which can also block depression‐related behaviors induced by PPM1F knockdown. Furthermore, inhibiting the acetylase activity of p300 abolished the beneficial effects of PPM1F elevation on CUS‐induced depressive behaviors. Conclusion Our findings demonstrate that PPM1F in the mPFC modulates depression‐related behavioral responses by regulating the function of p300 via the AMPK signaling pathway.


| INTRODUC TI ON
Depression is a common and debilitating psychiatric disorder, with core features of distinct and persistent loss of interest and low mood.Depression affects over 350 million people worldwide and can result in serious dysfunction in patients. 1,2Currently, although pharmacological therapies for major depressive disorder are available, the limitations of medical treatments for depression often show a high-rate ineffectiveness, and are concomitant with intolerable side effects. 3,4Thus, a deeper exploration of the pathogenic mechanisms of depression is urgently required to provide insights for the development of novel therapeutic drugs.
The medial prefrontal cortex (mPFC) is a cortical region with different cell types and projections that innervate numerous brain areas, and is recognized to play an essential role in intellectual emotional control, including in relation to anxiety and depression. 5Volumetric reductions in the mPFC are among the most well-documented neural abnormalities in patients with major depressive disorder. 6Further, protein phosphatase Mg2+/Mn2+dependent 1F (PPM1F) is a serine/threonine phosphatase from the protein phosphatase 2C family, 7,8 and its functional activity has been linked to apoptosis regulation, proliferation and metastasis of cancer cells. 9,10Previous research has identified six PPM1F single-nucleotide polymorphisms that affect the association between posttraumatic stress disorder symptom severity and cortical thickness of the bilateral superior frontal and orbitofrontal regions, suggesting that variants of this gene may be relevant to the neural integrity of the prefrontal cortex (PFC). 11Our previous research found that dysfunctional elevation of PPM1F expression in the hippocampus is a crucial factor affecting depression and anxiety, and bidirectional modulation of PPM1F expression in the dentate gyrus can produce diverse phenotypes associated with depression and anxiety. 12,13In addition, an association between dysfunctional levels of PPM1F in the mPFC and depression and anxiety has been revealed in both animals and humans. 14,15However, it remains unclear whether PPM1F in the mPFC is involved in the pathogenesis of major depression.p300 is a histone acetyltransferase (HAT) that acetylates histones and a growing list of transcription-related proteins to promote transcription activity by loosening chromatin, enhancing their DNA-binding activity, and facilitating protein-DNA recognition and protein-protein interactions. 16,17Further, p300 is reportedly linked to the pathogenesis of several cancer modalities and is critical for both object recognition and contextual fear memory, and neurodegenerative disorders. 16,18,19 Marek et al. reported that infusing a small-molecule p300-specific inhibitor (C646) into the infralimbic cortex of the PFC, in which p300 is highly expressed within pyramidal neurons, enhances the consolidation of fear extinction memory by accelerating long-term potentiation under weak extinction training. 20However, considering the important role of p300, limited research has been conducted on its function within the context of psychiatric disorders, especially for depression.Therefore, an extensive examination of the neural and cellular mechanisms of p300 is needed.
In this study, we first established a depression-related animal model induced by chronic unpredictable stress (CUS) and analyzed the gene expression levels of PPM1F in the mPFC in relation to depressive behaviors.We then used an adeno-associated virus strategy to determine the impact of knockdown or overexpression of PPM1F in the excitatory neurons on depression-related behaviors under basal and stress conditions in both male and female mice.In addition, we tested whether knockdown of PPM1F regulates the neuronal excitability of pyramidal neurons in the mPFC, and identified the causal relationship between neuronal excitability and depression-related behaviors.We also examined the expression levels of p300 after PPM1F knockdown, accompanied by abnormal AMPK phosphorylation levels, and tested the depression-related behaviors induced by PPM1F knockdown after AMPKα2 knockout.Finally, we investigated whether the acetylation activity of p300 was required for the antidepressant activity of PPM1F overexpression.
Award Number: 202003090720 and 202003070728; the Special Funds of Taishan Scholars Project of Shandong Province, Grant/Award Number: tsqn202211368 expression of CREB-binding protein (CBP)/E1A-associated protein (p300), a histone acetyltransferase (HAT), and induced hyperphosphorylation of AMPK, resulting in microglial activation and upregulation of proinflammatory cytokines.Conditional knockout of AMPK revealed an antidepressant phenotype, which can also block depression-related behaviors induced by PPM1F knockdown.Furthermore, inhibiting the acetylase activity of p300 abolished the beneficial effects of PPM1F elevation on CUS-induced depressive behaviors.

| Stereotaxic surgery, microinjection, and cannulation
The coding region of mouse PPM1F (NM_176833) with calcium/ calmodulin-dependent protein kinase II (CaMKII) promoter or the short hairpin RNA oligonucleotides GGATG AGA AAG CAC GAA TTGA and GCATA CCA ATG CTT CTC ACCA targeted PPM1F with an independent U6 promoter were packaged into an adeno-associated virus (AAV2/9) with titers >1 × 10 12 vg/mL (Hanbio, shanghai, China).The oligonucleotide TTCTC CGA ACG TGT CACGT was used as a nonspecific control (NC).The efficiency and specificity of overexpression or knockdown of PPM1F were confirmed using quantitative real-time PCR. 12l surgeries were conducted using a stereotaxic apparatus (Kopf Instrument) under deep anesthesia conditions using isoflurane.vectors (per side) were injected at a controlled rate of 100 nL/min using a mineral oil-filled glass micropipette with a UMP3 microsyringe pump (World Precision Instruments).Behavioral tests were conducted 3 weeks after viral injection.For all viral infections, animals with incorrect injection sites were excluded from data analysis.
For intra-mPFC microinjection of the C646, bilateral guide cannula (23-gauge; RWD Life Science CO., Ltd.) was implanted 1 mm above the mPFC (coordinates: AP = +1.78mm, ML = ± 0.4 mm, DV = −1.6 mm from the bregma) of adult male C57BL/6J mice (8week old) as previously described. 24Microinjections were performed on freely moving mice in their home cage.A bilateral injection cannula (33-gauge) connected to a 5-μL syringe was inserted into the guide cannula.C646 or vehicle was infused into the mPFC in a volume of 0.5 μL over 5 min using an infusion pump (KD Scientific Inc.).The injector tips were held in place for another more 5 min after the end of injection to avoid backflow.The mice with loosed or missed canulations in each group were not included in the statistical analysis.

| Immunofluorescence histochemistry analysis
Mice were deeply anesthetized and transcardially perfused with ice-cold PBS followed by 4% paraformaldehyde.The intact brains were fixed in 4% paraformaldehyde for 24 h at 4°C.After being washed with PBS, the brains were stored in a 30% sucrose solution for 48 h at 4°C.Frozen brains were sectioned at 40 μm thickness.

| Western blot analysis
Total proteins were extracted as previously described. 23,26,27The Fluorescence was visualized and analyzed using an Odyssey infrared imaging system (Li-COR Biosciences).

| Electrophysiological recordings
Electrophysiological recordings were conducted as previously described. 24First, the brains from the anesthetized mice with isoflurane were quickly transferred to an ice-cold solution (254 mM sucrose, The input resistance was measured using a hyperpolarizing current injection of −20 pA from a −80 mV holding current.The fast after hyperpolarization (fAHP) size was tested as the difference between the spike threshold and voltage minimum after the action potential peak.

| Chronic unpredictable stress and behavioral procedures
The CUS procedure was carried out as described previously with minor modification 28,29 to maximize the unpredictability and mildness of the stress intensity.Briefly, mice were subjected to two different stressors at different times of every day for 14 consecutive days, which included restraint (2-h), tail pinch (15-min), constant light (24-h), wet bedding (24-h) with 45° cage tilt, inescapable foot shocks (10-min, 0.3 mA, 2 s duration, the interval is 16 s), elevated platform (30-min), and social isolation. 30Control mice were group housed and briefly handled daily in the housing room.An abbreviated (7 days) subchronic unpredictable stress (SCUS) was conducted to assess stress susceptibility. 28l behavioral tests were performed during the late light phase of the lighting cycle, except for the sucrose preference test, which was conducted in the initial 2 h of the dark phase.Prior to each test, all mice were habituated to the testing room for 4 h.To avoid possible subjective effects, the behavioral performance of each mouse in each test was recorded using a double-blind procedure.To avoid possible carryover effects, the adjacent behavioral tests were spaced apart by 2-3 days.The procedures used for the behavioral tests are presented sequentially below.

| Female urine sniffing test
We used a female urine sniffing test (FUST) to examine the sexrelated reward-seeking behavior of male mice based on their interest in pheromonal odors from estrus female urine. 31As previously described, 12,26 estrus female urine was collected by monitoring the estrus cycle of female mice using microscopic examination of vaginal smears.Before testing, male mice were habituated to a sterile cotton-tipped applicator inserted into their home cage for 1 h before being transferred to a testing room with constant dim light (approximately 3 lux).The test was conducted in three stages: (1)   exposure to a cotton-tipped applicator dipped in sterile water for 3 min, (2) a 45-min interval, and (3) exposure to a cotton-tipped applicator dipped in fresh urine (80 μL) of estrus female mice for 3 min.
The sniffing times for sterile water and female urine were recorded respectively.

| Sucrose preference test (SPT)
As described previously, 28 1 week before testing, mice were habituated to drinking water from two tubes (50 mL) with stoppers fitted with ball-point sipper tubes.Four hours before testing, mice were transferred to the testing room from the group to individual housing without water.During testing, a free choice between a tube of either water or sucrose solution (1%) was presented to mice for 2 h, and the water and sucrose solution consumption were recorded.The preference for sucrose solution was calculated as the ratio of the consumption of sucrose solution to the total consumption of water and sucrose solution.

| Forced swim test (FST)
The apparatus of the forced swim test (FST) was a clear Plexiglas cylinder (10 cm in diameter × 25 cm in height) filled with warm water (24°C, 15 cm in depth).At the beginning of the test, the mice were placed individually into the cylinder for 6 min.The total duration of immobility was recorded during the last 4 min of the 6 min test using a video camera positioned directly above the cylinder.Immobility was defined as the absence of any body or limb movement except for those caused by respiration.

| Novelty suppressed food test (NSFT)
A novelty suppressed food test (NSFT) was conducted according to previously published protocols. 32,33Mice were deprived of food for 24 h.Testing was performed in a plastic box (60 × 60 × 40 cm 3 ) covered with bedding and illuminated with dim light.A single pellet of food on round filter paper (11 cm in diameter) was placed in the center of the box, and a mouse was placed in the corner.The latency period before eating was recorded for 10 min.The mice were then returned to their home cages.Food consumption was calculated after 5, 10, and 30 min.

| Elevated plus maze test (EPM)
According to the previous paper, 25

| Locomotor activity
The locomotor activity of the mice was assayed in SuperFlex Fusion open-field cages (40 × 40 × 30 cm, Omnitech Electronics Inc.).A single mouse was gently placed in the corner of the cage and allowed to explore the field freely under illuminated conditions for 30 min.Mouse movements were monitored and recorded using infrared motion sensors attached to the tops of the cages.The total distance traveled was quantified using Automated Fusion Software (Omnitech Electronics Inc.).

| Statistical analyses
All statistical analyses were performed using GraphPad Prism version 8.The normality and equal variance assumptions were tested using the Shapiro-Wilk and F tests.Two-tailed unpaired t-tests were used to assess the differences between the two experimental groups for normally distributed data.Two-tailed t-tests with

| Appearance of depression-related behaviors induced by PPM1F knockdown
To examine the role of PPM1F in depression, we packaged an adenoassociated virus mediating PPM1F knockdown using a short hairpin RNA (AAV-shRNA-GFP) strategy under the CaMKII promoter, which specifically targets excitatory neurons. 33Then, this virus was infused into the mPFC, and depression-related behaviors were measured after 3 weeks (Figure 2A).Fluorescence analysis revealed that the expression area of AAV-shRNA-GFP was limited in the mPFC, while the specificity and efficiency of the PPM1F targeting shRNA, which has been previously demonstrated, 12 was tested again to show that mRNA (t (12) = 4.8750, p < 0.0010) and protein (t (12) = 3.9740, p = 0.0018) levels were both significantly decreased compared to the control AAV-NC-GFP group (Figure 2B).Furthermore, we found that the sniffing time for water was nearly the same (p > 0.9999), but significantly decreased for female urine of the male AAV-shRNA-GFP mice (p = 0.0040; Figure 2C Mann-Whitney U test, p = 0.3278; female: Mann-Whitney U test, p = 0.6048).In the NSFT, there was an increased latency to food only in male AAV-shRNA-GFP mice (male: t (16) = 3.4580, p = 0.0032; female: t (16) = 0.0876, p = 0.9313), and none compared changes in the food intake of both male and female mice (Figure 2E, male: treatment:  S2B).We also found that male and female AAV-shRNA-GFP-injected mice showed no differences in the center time (male: t (16) = 0.7749, p = 0.4497; female: t (16) = 0.5068, p = 0.6192) or total distance (male: t (16) = 1.1890, p = 0.2518; female: t (16) = 0.9853, p = 0.3391) compared to AAV-NC-GFP-injected mice in the OFT (Figure S2C).

| Overexpression of PPM1F produces an antidepressant effect in depressed mice
Next, we packaged adeno-associated virus that mediates PPM1F overexpression under the CaMKII promoter.Depression-related behaviors were measured after injecting this virus into the mPFC (Figure S3A).We observed that GFP fluorescence was concentrated in the mPFC, and protein (t (10) = 2.3570, p = 0.0402) levels were increased in the AAV-PPM1F-GFP group compared with the control AAV-GFP group (Figure 3A).We also found that the sniffing times for both water (p > 0.9999) and female urine (p = 0.1968) were similar to those of male AAV-GFP mice (Figure S3B We observed no clear differences for sucrose preference  showed no significant changes in these groups (Figure S3F, male: treatment: We also evaluated anxiety-related behaviors (Figure S4A with AAV-GFP-injected mice in the EPM test (Figure S4B).We also found that the time in the center (male: t (17) = 0.8248, p = 0.4209; female: Mann-Whitney U test, p = 0.5787) and total distance (male: t (17) = 1.2320, p = 0.2349; female: t (18) = 0.4675, p = 0.6457) were not affected for either male or female AAV-PPM1F-GFP-injected mice compared with AAV-GFP-injected mice in the OFT (Figure S4C).
Furthermore, we measured the effects of PPM1F overexpression in the mPFC on depression-related behaviors under stress conditions by subjecting the mice to CUS (Figure 3A).The results showed that CUS significantly reduced sniffing time for female urine (p = 0.0093), which was consistent with our previous reports, 28,33 and that PPM1F overexpression can rescue this decline (p = 0.0283; Figure 3B,

| PPM1F knockdown decreased the excitability of pyramidal neurons in the mPFC
Electrophysiological dysfunction is a pivotal characteristic of the pathophysiological process of depression.Therefore, we recorded the excitability of neurons in the mPFC after PPM1F knockdown (Figure 4A).The results showed fewer action potentials in response to the same amount of current injections (Figure 4B

| Counteracting the reduced excitability of pyramidal neurons restores the depression-related behaviors induced by PPM1F knockdown
Next, we used a chemogenetic approach to determine whether the activation of blocked neuronal excitability of the neurons in the mPFC induced by PPM1F knockdown was sufficient to reverse depressionrelated behaviors.We infused the mixtures of AAV-shRNA-GFP, AAV-NC-GFP, AAV-hM3D-mCherry or AAV-mCherry viruses into the mPFC bilaterally, and then injected clozapine-N-oxide (CNO) to activate neuronal excitability of the neurons in the mPFC.After finding that the coinjected viruses were robustly coexpressed in the mPFC, we evaluated the depression-related behaviors (Figure 5A,B).
The results showed that AAV-hM3-mCherry mediating neural activation prevented decreases in sniffing time for female urine were similar between both groups.

| PPM1F regulated p300 expression by modifying AMPK activity
To verify the role of p300 in depression, we first measured the expression levels of p300 in the mPFC of depressed male and female mice, and found that CUS decreased the mRNA levels of p300 only in male mice (Figure 6A, male: t (18) = 3.0580, p = 0.0068; female: t (14) = 0.1554, p = 0.8787).Furthermore, only the p300 mRNA levels of male mice were positively correlated with depression-related behaviors, such as the sucrose preference (Figure 6B).We then tested whether p300 expression was regulated by PPM1F, and observed that PPM1F knockdown significantly decreased the mRNA levels of p300 (Figure 6C t (12) = 4.0400, p = 0.0016), and p300 mRNA levels were positively correlated with the mRNA or protein levels of PPM1F (Figure 6C).Previous research has reported that AMPK is a downstream signaling pathway of PPM1F, 12 and the activity of p300 is inhibited by AMPK mediating phosphorylation. 34Therefore, we measured the effect of PPM1F on the activity of AMPK and found that the phosphorylation levels of AMPK were obviously increased by PPM1F knockdown (t (12) = 3.6230, p = 0.0035) and positively correlated with the mRNA and protein levels of PPM1F (Figure 6D).
Next, as many reports have suggested a significant association between neuroinflammation and depression, 35,36 and our previous research has implied that PPM1F may regulate the neuroinflammation reactivity, 37 therefore, we investigated the neuroinflammationrelated molecules after PPM1F knockdown in the mPFC, we found that the number of iba1 + cells, a marker for microglia was increased in the PPM1F knockdown group (Figure S6A, t (10) = 2.7850, p = 0.0193).
Meanwhile, the levels of iNOS and Arg-1, markers of microglia/ macrophage polarization were measured, and the results indicated that the mRNA level of iNOS remains unchanged (t (12) = 1.0870, p = 0.2986), while the mRNA level of Arg-1 was decreased in the PPM1F knockdown group (t (12) = 3.4010, p = 0.0053; Figure S6B).

| Knockout of AMPK produced an antidepressant activity under stress conditions
Based on the finding that PPM1F knockdown increased the phosphorylation levels of AMPK, which can produce depressiverelated behaviors, we assumed that knockout of AMPK may bear the antidepressant activity.Therefore, we first injected the AAV-CaMKII-GFP or AAV-CaMKII-Cre virus into the mPFC of AMPKα2 flox mice to knock out the α2 subunit of AMPK, which was the targeted phosphorylation site of AMPK (Figure S7A).The fluorescence results showed that the infused virus was specifically expressed in the mPFC, and the mRNA levels of exon2 of AMPK dramatically decreased (Figure S7B, (t (10) = 2.6300, p = 0.0252).
Next, we measured depression-related behaviors without stress and found that AAV-CaMKII-Cre-treated mice showed nearly the AAV-CaMKII-Cre and AAV-CaMKII-GFP control mice in the EPM test (Figure S8B).Moreover, we found that the time in the center (t (16) = 1.0710, p = 0.2999) and total distance (t (16) = 1.2770, p = 0.2199) did not differ in AAV-CaMKII-Cre-treated mice compared with AAV-CaMKII-GFP mice in the OFT (Figure S8C).

| AMPK knockout blocked depression-related behaviors induced by PPM1F knockdown
To identify whether AMPK is a pivotal and integral mediator in the relationship between PPM1F and depression, we further investigated analysis showed that the infused virus was coexpressed in the mPFC (Figure 7A).In the FUST, an decreased sniffing time for female urine (p = 0.0338) induced by AAV-shRNA-GFP treatment was ameliorated by AAV-CaMKII-Cre-mCherry treatment (p = 0.0046;

| Regulation of depression-related behaviors and stress sensitivity by inhibiting the acetylase activity of p300
Acetylase activity is the key functional component of p300; 16,23 therefore, we further explored whether the acetylase activity of p300 could be involved in inducing depression-related behaviors.
C646, 38 a competitive p300 histone acetyltransferase, was applied to the mPFC of WT mice once daily for five days, and then the depression-related behaviors were evaluated (Figure S9A).

| Inhibiting the acetylase activity of p300 abolished the beneficial effects of PPM1F elevation in CUS-exposed mice
To examine whether the acetylase activity of p300 mediates the antidepressant activity induced by PPM1F overexpression in the mPFC, AAV-PPM1F-GFP or control virus AAV-GFP was injected into the mPFC, and these mice were then canulated 3 weeks later.Then, 1 week later, C646 was infused into the mPFC once every 2 days for 14 days.These mice were exposed to CUS, and their depressionrelated behaviors were evaluated (Figure 8A).

| DISCUSS ION
In the present study, we found that the expression of PPM1F was enclosed in the excitatory pyramidal neurons in the mPFC, and the abnormal expression of PPM1F played a pivotal role in the process of depression.shRNA-mediated genetic knockdown of PPM1F in the mPFC produced depression-related behaviors, while overexpression of PPM1F alleviated these behavioral responses in CUS-exposed mice.Moreover, PPM1F exerted a regulatory function by affecting the excitability of pyramidal neurons in the mPFC and controlled the expression of p300 via the hyperphosphorylation of AMPK.
Determination of pathological genes is a direct and critical method of identifying the path for understanding depression or other mental disorders. 39Our previous studies revealed that significant upregulation of PPM1F in the hippocampus is related to the behavioral and pathological states of depression and anxiety. 12,13wever, in the current study, it was interesting to find that the expression levels of PPM1F in depressed mice decreased in the mPFC, another morbigenous region underlying the pathophysiology of depression, 2,6,40 and this result was consistent with the another research that the reduced expression levels of PPM1F in mPFC was observed in the mice subjected to acute immobilization stress. 14rthermore, downregulation of PPM1F in the mPFC increased depressive behaviors, validating the causal relationship between PPM1F and depression.Another critical founding was that the only depressed male mice displayed the reduced expression levels of PPM1F in mPFC, demonstrating the gender difference phenomenon, which was in line with the reported results that PPM1F protein levels and colocalization with CAMKIIG were altered in mPFC of only male mice after immobilization stress. 14However, inconsistent with our previous studies, the levels of PPM1F in the hippocampus was increased of depressed mice of both sexes, and was elevated only in female anxious mice. 12,13The above results indicated the mixed gender and brain region specific characteristics of dysfunctional levels of PPM1F in depression. 41In addition, there were also slightly sex differences in the depression-or anxiety-related behaviors after PPM1F knockdown in mPFC, with more prominent phenotypes in male than female, the anxious behaviors was obvious in females than in males, while the anxiolytic activity was apparent in males than in females.There are some possible explanations for these characteristics, such as the existing sex differences in corticosterone levels following CUS, 42 and corticosterone was reported to regulate the expression of PPM1F. 14Moreover, the estrogen signaling system has diverse effects in the morphology and activity of mPFC neurons. 43e regulated molecular pathways and epigenetic modifications by PPM1F also vary between the sexes. 13normal neuronal excitability has been well recognized as one magnified characteristic of the etiology and pathogenesis of depression. 2,44Notably, the exact role of the neuronal excitability of the mPFC pyramidal neurons in depression was ambiguous, because some reports from our group and another group indicate these neurons show a hypoactive state in depression, 24,45 while the hyperexcitable state of these neurons in depression has also been recorded. 46,47These contradictory results may be related to the existing fact for region-or projection-specific characteristics of these neurons.9][50] In the present study, lower neuronal excitability in the pyramidal neurons in the mPFC was also observed in the excitatory neural-specific PPM1F knockdown mice, while restoring this hypoactive condition alleviated the depressive phenotypes induced by PPM1F knockdown.Interestingly, unlike the none altered levels of hippocampal kcnn1 and kcnf1 after PPM1F overexpression, 12 we observed that the expression levels of kcnn1 and kcnf1 were suppressed after PPM1F knockdown, which may mediate the hyperexcitability induced by PPM1F knockdown. 51erefore, our results indicate that the hypoactive condition of pyramidal neurons is a key factor in the pathological process of depression, but the exact roles of kcnn1 and kcnf1 in affecting the neural excitability still need further determination.
AMPK is a serine/threonine kinase that functions as a key energy sensor in a wide variety of tissues and as a neuroprotective protein in the brain. 52Some research conducted by our group and others has reported that AMPK can be dephosphorylated by PPM1F or other protein phosphatase 2C family proteins.studies have shown that rosiglitazone or metformin can exert an antidepressant effect in unpredictable stress-induced depressive mouse models by activating the AMPK signaling pathway, 54,55 and directly igniting AMPK itself by activating the 5-aminoimidazole -4-carboxamide-1β-d-ribonucleotide (AICAR) also produces antidepressant effects. 56However, a systematic understanding of AMPK dysfunction in depression is currently lacking in the literature.We found that PPM1F knockdown in the mPFC can result in the activation of AMPK with an elevated level of p-T172, which is consistent with the inhibitory effect of PPM1F overexpression on AMPK activation in the hippocampus. 12These results suggest that AMPK may be the underlying signaling protein of PPM1F in depression.Further investigation revealed that specific knockout of AMPKα2, in which subunit the phosphorylation site of threonine 172 is localized, produced antidepressive behaviors.Moreover, we also demonstrated that AMPKα2 mediates the depressive behaviors induced by PPM1F knockdown in the mPFC.To our knowledge, our study is the first to demonstrate the function of AMPK in the emergence of depression, and it also displayed as a precious participant in the role of PPM1F in depression.The unexpected and notable results in this study were the opposite modulation of PPM1F expression in the mPFC of CUS treated mice compared to the hippocampal expression levels in our previous reports, 12 with the discrepant role in affecting depression-related behaviors and underlying mechanism.In detail, CUS reduced the expression levels of PPM1F in mPFC, while elevated its levels in the hippocampus, knockdown of PPM1F in mPFC induced depression-related behaviors, but produced antidepressant activity in hippocampus, accompanied with the different and individual effect on the neural excitability and signaling pathways.Nevertheless, the concrete reasons remain unclear and intricate, the diverse neural circuit connections and subregion-specific or rostral-caudal axis-specific reactivity to stress may account for this divergence, 2,57-60 which needs further exploration in the future.
We placed specific focus on acetylase p300 based on our previous research, which showed that p300 could mediate the antidepressant effect of leptin by regulating the expression of BDNF through histone H3 acetylation. 23In addition to its present role in serving as the presence of histone acetyltransferase activity, which endows p300 with the capacity to modulate structure-sensitive chromatin with histone acetylation modifications, p300, as a transcriptional coactivator protein, plays a central role in coordinating and regulating many signal-dependent events with the transcription process and other multiple ways. 16For example, it acts as a protein bridge, thereby connecting diverse transcription factors to transcription sites and also provides a protein scaffold for the formation of a multicomponent transcriptional regulatory complex.However, the exact role and upstream molecular modifier(s) of p300 remain unclear.Thus, AMPK can directly phosphorylate and downregulate p300 mediated acetylation and transcription functions, 34,61 which suggests that AMPK may be a potential modulator of p300 activity.
Therefore, we raised the hypnosis that depression was developed from the downregulation of PPM1F, which enables the activation of AMPK, leading to the blocking effect of the p300-associated transcriptional promoter.Meanwhile, our results validated the hypothesis that decreased p300 expression levels occurred after PPM1F knockdown, which was abolished by AMPK knockout.However, an unexpected finding was that the infusion of C646, which is a smallmolecule specific inhibitor of p300, to the mPFC, could not induce depression-related behaviors in either the basal or subthreshold CUS conditions, but partially blocked the antidepressive effect of PPM1F overexpression.These features have some potential explanations, the first one we must remain is that the functional dimorphic characters of the subregions of mPFC, prelimbic (PrL) and infralimbic (IL) of the mPFC; 24,50,62,63 therefore, the infusion strategy might be improved to promote the location accuracy.Another possible explanation is that complicated types of neurons, pyramidal neurons, inhibitory GABAergic interneurons and excitatory interneurons coexist in this region; thus, the infused C646 may have a comprehensive effect.Furthermore, p300 is a nuclear phosphoprotein, and apart from AMPK, many kinases have been implicated in the phosphorylation process. 16,64The exact role of p300 in depression also awaits further elucidation using brain regions or neuralspecific manipulative strategies in the future.
In summary, this study's results indicate that PPM1F-AMPK-p300 activity in the mPFC neurons is a key regulator of depression-related behaviors, and our data suggest that abnormal PPM1F activity, impaired neuronal excitability in the pyramidal neurons and enhanced neuroinflammation state may contribute to the development of depression.However, how chronic stress causes a change in PPM1F expression in the mPFC, the neuronal populations and targeted genes of p300 mediating regulation of the effects of PPM1F on depression require further exploration.

:
Our findings demonstrate that PPM1F in the mPFC modulates depressionrelated behavioral responses by regulating the function of p300 via the AMPK signaling pathway.K E Y W O R D S depression-related behaviors, mPFC, neuroinflammation, neuronal excitability, PPM1F-AMPK-p300 axis 2 | MATERIAL S AND ME THODS 2.1 | Animals Wild-type (WT) C57BL/6J mice (Stock No. 000664) and AMPKα2 flox/+ mice (Stock No. 014142), 21 which possess loxP sites flanking the region encoding exon 2 of AMP-activated protein kinase (AMPK), were purchased from Jackson Laboratory (Bar Harbor, ME, USA).All the mice were backcrossed and maintained to a C57BL/6J background.Heterozygous flox/+ transgenic mice were crossed to produce homozygous offsprings.For genotyping, the following PCR primers were used: AMPKα2 WT and flox: forward-5′-GCAGGC GAATTTCTGAGTTC-3′, reverse-5′-TCCCC TTG AAC AAG CATACC-3′.The animals were housed in groups of 3-5 per cage under a 12 h light/dark cycle (lights on at 7:00 a.m.) with water ad libitum and standard food pellets.Both male and female mice (8-12 weeks old) were used.This study was approved by the Institutional Animal Care and Use Committee of Binzhou Medical University Hospital 22 and conducted in accordance with the U.K. Animals (Scientific Procedures) Act, 1986 and associated guidelines, EU Directive 2010/63/EU for animal experiments.

2 . 8 . 6 |
the maze consisted of four arms arranged around a central platform (5 × 5 cm 2 ) that with open access to any arm with white Plexiglas.Testing was conducted under bright white light.The test was initiated by lacing each mouse in the central area of the maze facing the corner between a closed arm and an open arm, and allowed to explore the elevated plus maze for 5 min.The maze was thoroughly cleaned with 20% ethanol after each trial.The time spent on the open and closed arms and the number of entries into each arm were scored.The anxiety-related behavior was evaluated by calculating the percentage of open arm time (time spent in the open arms / total time spent in all arms) and the percentage of open arm entries (entries into the open arms / total entries into all arms).Open field test (OFT) An open field test (OFT) was performed in a nontransparent square box with an open area of 60 × 60 cm 2 and walls 40 cm in height.The entire test area was adjusted to even illumination with a house light on the ceiling.At the beginning of each session, mice were placed in the center of the open field area, and their activity was recorded for 5 min.The arena was thoroughly cleaned with 20% ethanol between tests.For the analysis, the open field arena was divided into nine equal squares using a 3 × 3 grid.The assessed parameters were time spent in the center zone and total distance traveled, quantified using Any-maze software (Stoelting).

Welch's correction
were used to analyze the normally distributed data with unequal variances, whereas Mann-Whitney U tests were used for non-normally distributed data.For the analysis of the three groups, one-way analyses of variance (ANOVAs) followed by Sidak post hoc tests were used for normally distributed data.The Kruskal-Wallis test followed by Dunn's multiple comparisons test was used for non-normally distributed data.For multiple groups, two-way or two-way repeated-measures ANOVAs were used followed by Tukey's test.The linear relationships between two variables were analyzed by calculating Pearson's correlation coefficient, the 95% confidence bands were indicated by dotted lines.Statistical significance was set at p < 0.05.All data are presented as means ± standard errors (SEM).A Grubbs outlier test was performed, and samples that varied by >2 standard deviations from the mean were removed.