Upregulation of AMPA receptor GluA1 phosphorylation by blocking adenosine A1 receptors in the male rat forebrain

Abstract Objective The adenosine A1 receptor is a Gαi/o protein‐coupled receptor and inhibits upon activation cAMP formation and protein kinase A (PKA) activity. As a widely expressed receptor in the mammalian brain, A1 receptors are implicated in the modulation of a variety of neuronal and synaptic activities. In this study, we investigated the role of A1 receptors in the regulation of α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic acid (AMPA) receptors in the adult rat brain in vivo. Methods Adult male Wistar rats were used in this study. After a systemic injection of the A1 antagonist DPCPX, rats were sacrificed and several forebrain regions were collected for assessing changes in phosphorylation of AMPA receptors using Western blots. Results A systemic injection of the A1 antagonist DPCPX induced an increase in phosphorylation of AMPA receptor GluA1 subunits at a PKA‐dependent site, serine 845 (S845), in the two subdivisions of the striatum, the caudate putamen, and nucleus accumbens. DPCPX also increased S845 phosphorylation in the medial prefrontal cortex (mPFC) and hippocampus. The DPCPX‐stimulated S845 phosphorylation was a transient and reversible event. Blockade of Gαs/olf‐coupled dopamine D1 receptors with a D1 antagonist SCH23390 abolished the responses of S845 phosphorylation to DPCPX in the striatum, mPFC, and hippocampus. DPCPX had no significant impact on phosphorylation of GluA1 at serine 831 and on expression of total GluA1 proteins in all forebrain regions surveyed. Conclusion These data demonstrate that adenosine A1 receptors maintain an inhibitory tone on GluA1 S845 phosphorylation under normal conditions. Blocking this inhibitory tone leads to the upregulation of GluA1 S845 phosphorylation in the striatum, mPFC, and hippocampus via a D1‐dependent manner.


S IG NIFIC AN CE
In this study, we found that an A 1 receptor antagonist DPCPX increased phosphorylation of AMPA receptor GluA1 subunits in the key limbic reward regions, including the striatum, medial prefrontal cortex, and hippocampus. Blocking dopamine D 1 receptors abolished the responses of GluA1 phosphorylation to DPCPX. These data reveal an inhibitory linkage from adenosine A 1 receptors to AMPA receptors. Malfunction of this linkage may be linked to various psychiatric and neurological disorders such as drug addiction and anhedonic depression.
In this study, we attempted to evaluate the role of A 1 receptors in the regulation of GluA1 phosphorylation and expression in the adult rat brain in vivo. To achieve this, we investigated the effect of an A 1 antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) on GluA1 phosphorylation at S845 and S831 sites in different brain regions, including the caudate putamen (CPu), nucleus accumbens (NAc), medial prefrontal cortex (mPFC), and hippocampus. We then tested the effect of DPCPX on GluA1 phosphorylation in the presence of a dopamine D 1 antagonist SCH23390.

| Animals
We used male rats in this study. Adult Wistar rats (2-3 months, 310-380 g) were purchased from a vendor (Charles River) and were kept at 12-hr light/12-hr dark cycle and 23°C with water and food available ad libitum. At 5-6 days after habituation, animals were used for experiments. The protocol of animal use and care was approved by the Institutional Animal Care and Use Committee.

| Drug administration and experimental arrangements
All drugs were administered via intraperitoneal (i.p.) injections. We calculated doses of drugs as their salt form. All drugs were injected in a volume of 1 ml/kg. Age-matched rats were injected with a vehicle solution and served as controls. A series of experiments were conducted with rats randomly divided into different groups using a computer-generated randomization table (GraphPad software/ QuickCalcs). Size of sample was determined by the sample size calculation (α = .05, β = .2; 80% power). There were no differences in sample size between the beginning and end of the experiments.
We based the criteria for inclusion/exclusion on the health state of animals. The healthy animals that showed no sign of illness as evaluated by the body weight and visual observations were used in the analysis.
We first investigated the effect of DPCPX, an A 1 antagonist which was approximately 1,000-fold more potent at A 1 than A 2A receptors in the striatum (Fredholm & Lindstrom, 1999). Three groups of rats (n = 4 per group) were subjected to an i.p. injection of vehicle or DPCPX at either 0.25 or 2.5 mg/kg and were then sacrificed 20 min after DPCPX injection. Changes in GluA1 phosphorylation and expression in four different brain regions, including the CPu, NAc, mPFC, and hippocampus, were subsequently assayed using Western blots. Doses of DPCPX (0.25 and 2.5 mg/kg) were selected because an i.p. injection of DPCPX at 3 mg/kg blocked the anxiolytic-like response to the A 1 selective agonist and positive allosteric modulator (Prediger, Silva, Batista, Bittencourt, & Takahashi, 2006;Vincenzi et al., 2016). Second, we conducted a time course study.
Following an injection of vehicle or DPCPX (2.5 mg/kg, i.p.), rats were sacrificed at different time points (1, 3, and 6 hr) for analyzing changes in GluA1 phosphorylation and expression in different brain regions. Two groups of rats (n = 4 per group) were used at each time point. Finally, the effect of DPCPX was tested in the presence of a dopamine D 1 receptor antagonist in four groups of rats (n = 4 per group). The D 1 antagonist SCH23390 (0.1 mg/kg, i.p.) was administered 10 min prior to DPCPX (2.5 mg/kg, i.p.). Rats were sacrificed 15-20 min after DPCPX injection for immunoblot analysis of changes in GluA1 phosphorylation in different brain regions.
Homogenates were centrifuged at 800 g (10 min). Protein concentration of the supernatant was determined. Samples were then used for Western blot which was performed as described previously (Mao et al., 2013;Mao, Faris, & Wang, 2018). Briefly, on NuPAGE Novex 4%-12% gels (Invitrogen), proteins (32 µg/well) in brain lysates were separated. Separated proteins on gels were transferred to membranes (polyvinylidene fluoride). The membranes were blocked, followed by incubation with a primary antibody overnight at 4°C. A secondary antibody linked to horseradish peroxidase was incubated to interact with the primary antibody. Immunoblots on membranes were visualized using an enhanced chemiluminescence reagent. To quantitatively analyze immunoblots, we measured optical density of blots using analysis software (NIH ImageJ, RRID: nif-0000-30467).
All values of optical density were normalized to β-actin.

| Antibodies
All primary antibodies used in this study and their characterizations are listed in Table 1. The rabbit antibodies against the phosphoryl-

| Statistics
Data were statistically analyzed following tests for the normality of data. They are expressed as means ± SEM. In this study, no test for outliers was conducted on the data. No rats were excluded from the analysis. As appropriate, a two-tailed unpaired Student's t test or one-or two-way analysis of variance (ANOVA) with a multiple comparison post hoc test was conducted to analyze data. The statistical significance was determined by a p value < .05.

| Effects of the A 1 antagonist DPCPX on GluA1 phosphorylation
To determine the impact of blockade of A 1 receptors on GluA1 phosphorylation, we subjected rats to a single i.p. injection of the A 1 antagonist DPCPX. We sacrificed rats 20 min after DPCPX injection to assay changes in GluA1 phosphorylation at S845 and S831 in different brain regions using Western blots. In the CPu, DPCPX at a lower dose (0.25 mg/kg) had no significant effect on GluA1 phosphorylation at S845 and S831 ( Figure 1a). Noticeably, at a higher dose (2.5 mg/kg), DPCPX induced a marked increase in pS845 levels, while DPCPX did not alter pS831 levels. Cellular levels of total GluA1 proteins remained stable in response to DPCPX. Similar results were observed in the NAc. As shown in Figure  DPCPX. After an injection of DPCPX at 2.5 but not at 0.25 mg/kg, the pS845 level in the hippocampus was elevated in DPCPX-treated rats compared to vehicle-treated rats (Figure 2b). Little change in pS831 and total GluA1 levels was seen in the two regions following DPCPX administration at either dose. These data indicate that DPCPX was able to enhance S845 phosphorylation in the mPFC and hippocampus.

| A time course study of the DPCPX effect
We next carried out a time course study to define temporal properties of the DPCPX effect. To this end, we injected vehicle or DPCPX at an effective dose (2.5 mg/kg) to rats. We then sacrificed rats at different time points (1, 3, and 6 hr) after drug injection and analyzed changes in S845 and S831 phosphorylation in different brain regions. Like the positive effect of DPCPX observed at 20 min from the above studies, DPCPX at 1 hr elevated pS845 levels in the CPu ( Figure 3a). This elevation became insignificant at 3 hr. At 6 hr, no significant difference in pS845 levels was found between DPCPX-and vehicle-treated groups. In the NAc, a significant increase in pS845 levels was seen at 1 hr in DPCPX-treated rats compared to vehicletreated rats (Figure 3b). This increase persisted at 3 hr and declined to a level insignificantly different from the value obtained from the vehicle group at 6 hr. At all time points surveyed, DPCPX induced minimal changes in pS831 and GluA1 levels in the CPu and NAc.
Thus, DPCPX generally induced a short-lived and reversible increase in GluA1 S845 phosphorylation in the striatum.
We also examined the effect of DPCPX at different time points in the mPFC and hippocampus. In the mPFC, pS845 levels were substantially enhanced at 1 hr following DPCPX administration ( Figure 4a). Noticeably, pS845 levels in this region were reduced at 3 hr in DPCPX-treated rats relative to vehicle-treated rats. At 6 hr, an insignificant change in pS845 levels was induced by DPCPX.

F I G U R E 2
The effect of DPCPX on GluA1 phosphorylation in the rat mPFC and hippocampus. (a) The effect of DPCPX on GluA1 phosphorylation and expression in the mPFC. (b) The effect of DPCPX on GluA1 phosphorylation and expression in the hippocampus (Hippo). Note that DPCPX at a higher dose (2.5 mg/kg) but not a lower dose (0.25 mg/kg) induced an increase in pS845 levels in the mPFC and hippocampus. Representative immunoblots are shown left to the quantified data. Rats were administered with vehicle or DPCPX at either 0.25 or 2.5 mg/kg (i.p.) and were then sacrificed 20 min after drug administration for subsequent analysis of changes in S845 and S831 phosphorylation using Western blots. Data are presented as means ± SEM (n = 4 per group) and were analyzed using one-way ANOVA: mPFC-pS845, F(2,9) = 4.92, p = .036; mPFC-pS831, F(2,9) = 0.17, p = .844; mPFC-GluA1, F(2,9) = 1.38, p = .300; hippocampus-pS845, F(2,9) = 5.27, p = .031; hippocampus-pS831, F(2,9) = 0.22, p = .804; and hippocampus-GluA1, F(2,9) = 0.54, p = .599. *p < .05 versus vehicle Thus, it appears that DPCPX induces a biphasic pattern of changes in S845 phosphorylation in the mPFC, that is, an initial increase followed by a decrease in pS845 levels. In the hippocampus, DPCPX stimulated S845 phosphorylation at 1 and 3 hr, while DPCPX did not at 6 hr (Figure 4b), indicating that DPCPX induces a dynamic and reversible increase in S845 phosphorylation in the hippocampus.
Additionally, DPCPX had no significant effect on S831 phosphorylation and GluA1 expression in the mPFC and hippocampus at three time points surveyed.

| Effects of the D 1 antagonist on the DPCPXinduced GluA1 phosphorylation
To determine whether blockade of the dopamine D 1 receptor has any effect on the DPCPX-induced GluA1 phosphorylation, we

| D ISCUSS I ON
We initiated an effort to investigate the role of adenosine A 1 receptors in the regulation of GluA1 phosphorylation and expression in the adult rat forebrain in vivo. We found that a systemic injection of the A 1 antagonist DPCPX induced a significant increase in GluA1 phosphorylation at S845 in multiple brain regions, including the CPu, NAc, mPFC, and hippocampus. The increase in GluA1 phosphorylation induced by DPCPX was time-dependent and reversible. Pretreatment with a dopamine D 1 antagonist SCH23390 blocked the response of S845 phosphorylation to DPCPX in the striatum, mPFC, and hippocampus. In contrast to S845, GluA1 phosphorylation at S831 was not altered by DPCPX in all four brain regions. These data indicate that A 1 receptors exert tonic inhibition of GluA1 S845 phosphorylation in striatal, cortical, and hippocampal neurons under normal conditions. Blocking this tonic inhibition induces the D 1 -dependent upregulation of S845 phosphorylation.
Dopamine D 1 receptors are selectively expressed in striatonigral output neurons in the striatum (Aubert et al., 2000;Bertran-Gonzalez et al., 2010;Gerfen et al., 1990). Activation of G αs/ olf -coupled D 1 receptors stimulates the cAMP/PKA pathway, which in turn regulates a number of downstream effectors, including Note that DPCPX induced a biphasic effect in the mPFC, that is, an increase followed by a decrease in pS845 levels.
Representative immunoblots are shown left to the quantified data. Rats received an i.p. injection of vehicle or DPCPX (2.5 mg/kg) and were then sacrificed at a different time point (1, 3, or 6 hr) after drug administration for subsequent analysis of changes in S845 and S831 phosphorylation using Western blots. Data are presented as means ± SEM (n = 4 per group) and were analyzed using Student's t test: mPFC-pS845: p = .006 (1 hr), p = .022 (3 hr), and p = .522 (6 hr); and hippocampus-pS845: p = .001 (1 hr), p = .040 (3 hr), and p = .220 (6 hr GluA1 at a PKA-catalyzed phosphorylation site, that is, S845 (Chao et al., 2002;Mao et al., 2013;Price et al., 1999;Snyder et al., 2000;Swayze et al., 2004;Xue et al., 2014Xue et al., , 2017. In contrast, blockade of D 1 receptors resulted in a decrease in constitutive S845 phosphorylation in the striatum (Xue et al., 2017;this study). In addition to D 1 receptors, adenosine A 1 receptors are notably expressed in striatonigral neurons (Fuxe et al., 2007(Fuxe et al., , 2010. Since A 1 receptors are coupled to G αi/o proteins, activation of these receptors inhibits the cAMP/PKA pathway, whereas blockade of them leads to an opposite change in the cAMP/PKA pathway. As a downstream target of PKA, S845 phosphorylation is likely regulated by the A 1 tone. This study provides evidence in favor of this notion. We found that blocking A 1 receptors with DPCPX upregulated striatal S845 phosphorylation. This seems to support that there exists a tonic inhibitory tone from A 1 receptors, which inhibits basal S845 phosphorylation in striatal neurons. Moreover, given the colocalization of A 1 and D 1 receptors in striatonigral neurons, an A 1 -D 1 interaction model has been suggested. That is, two receptors form a balance to modulate the PKA signaling, thereby controlling the outflow of the D 1 /A 1 -mediated direct pathway in the basal ganglia (Ferre et al., 1994;Fuxe et al., 2007Fuxe et al., , 2010. Indeed, we found that blocking D 1 receptors abolished the A 1 antagonist-induced S845 phosphorylation, indicating that the D 1 /A 1 balance controls S845 phosphorylation. Others found that the A 1 agonist induced a late decrease in phosphorylation of dopamine-and cAMP-regulated phosphoprotein of M(r) 32 kDa at T34 via postsynaptic A 1 receptors in striatonigral neurons (Yabuuchi et al., 2006). DPCPX and the adenosine receptor antagonist caffeine elevated expression of c-fos and other immediate early genes and SCH23390 abolished the effect of caffeine in striatonigral neurons (Dassesse, Vanderwinden, Goldberg, Vanderhaeghen, & Schiffmann, 1999). Future studies will elucidate whether the D 1 /A 1 -regulated S845 phosphorylation occurs specifically in the phenotype of striatonigral output neurons. In addition to the postsynaptic D 1 /A 1 interaction, blocking presynaptic A 1 receptors by DPCPX may facilitate local dopamine release (Borycz et al., 2007;Wood, Kim, Boyar, & Hutchison, 1989). The released dopamine may stimulate D 1 receptors in striatonigral neurons to upregulate GluA1 S845 phosphorylation in these neurons.
A 1 receptors are expressed in the mPFC with a high level in pyramidal spiny output neurons (Rivkees, Price, & Zhou, 1995).
Ultrastructural analysis reveals the presence of A 1 receptors in both the presynaptic terminals and the postsynaptic structures (Ochiishi, Chen, et al., 1999 neurons and mossy fibers, but not on glial cells (Ochiishi, Chen, et al., 1999;Rivkees et al., 1995). At the subsynaptic level, A 1 receptors are distributed pre-and postsynaptically (Ochiishi, Chen, et al., 1999;Rebola, Pinheiro, Oliveira, Malva, & Cunha, 2003), indicating the importance of A 1 receptors in modulating both presynaptic neurotransmitter release and postsynaptic physiology. To determine the role of A 1 receptors in regulating GluA1 phosphorylation in hippocampal neurons, we found that the A 1 antagonist upregulated S845 phosphorylation in the adult rat hippocampus in vivo. Early studies found that A 1 receptors physically interacted with GluA1 subunits in hippocampal neurons (Chen et al., 2014). In rat hippocampal slices, prolonged A 1 receptor activation with CPA reduced GluA1 phosphorylation at S845 and S831 and induced GluA1 endocytosis and synaptic depression (Chen et al., 2014;Stockwell et al., 2016). These data suggest the hippocampus as another brain region where a significant inhibitory tone of A 1 receptors on S845 phosphorylation exists. Moreover, dopamine D 1 receptors are critical for the A 1 -mediated inhibition of S845 phosphorylation in this area. Pretreatment with the D 1 antagonist blocked the A 1 antagonist-induced S845 phosphorylation in the hippocampus.
Changes in S845 phosphorylation are thought to have a significant impact on the subcellular expression pattern and function of F I G U R E 6 The effect of SCH23390 on DPCPX-induced GluA1 S845 phosphorylation in the rat mPFC and hippocampus. (a) The effect of SCH23390 on DPCPX-induced S845 phosphorylation in the mPFC. (b) The effect of SCH23390 on DPCPX-induced S845 phosphorylation in the hippocampus (Hippo). Note that SCH23390 completely blocked the DPCPX-induced S845 phosphorylation in the two regions. Representative immunoblots are shown left to the quantified data. Rats were administered with SCH23390 (SCH, 0.1 mg/kg, i.p.) 10 min prior to DPCPX (2.5 mg/kg, i.p.) and were then sacrificed 15-20 min after DPCPX administration for subsequent analysis of changes in S845 and S831 phosphorylation using Western blots. Data are presented as means ± SEM (n = 4 per group) and were analyzed using two-way ANOVA: mPFC-pS845: saline versus SCH23390, F(1,12) = 33.01, p < .001, vehicle versus DPCPX, F(1,12) = 4.14, p = .064, and interaction, GluA1-containing AMPA receptors. Previous studies found that S845 phosphorylation enhanced the surface trafficking of GluA1 (Man, Sekine-Aizawa, & Huganir, 2007;Oh, Derkach, Guire, & Soderling, 2006) and AMPA channel peak current (Roche et al., 1996). Since the AMPA receptor-mediated glutamatergic transmission plays a pivotal role in the regulation of neuronal and synaptic activities in all the forebrain regions surveyed (striatum, mPFC, and hippocampus), the A 1 receptor-regulated S845 phosphorylation is believed to be a biochemical step critical for carrying out normal functions of these brain structures, a topic to be investigated in-depth in the future.
Our studies on AMPA receptor phosphorylation were performed in male rats. While sex differences in the A 1 -regulated AMPA receptor phosphorylation have not been reported in striatal neurons to our knowledge, it is important to assess the response of AMPA receptors to adenosine receptor agents in female animals in future studies.

ACK N OWLED G M ENTS
This work was supported by an NIH grant R01MH61469 (JQW). We thank Daozhong Jin and Nan He for their assistance.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.