Expression of p53 Target Genes in the Early Phase of Long-Term Potentiation in the Rat Hippocampal CA1 Area

Gene expression plays an important role in the mechanisms of long-term potentiation (LTP), which is a widely accepted experimental model of synaptic plasticity. We have studied the expression of at least 50 genes that are transcriptionally regulated by p53, as well as other genes that are related to p53-dependent processes, in the early phase of LTP. Within 30 min after Schaffer collaterals (SC) tetanization, increases in the mRNA and protein levels of Bax, which are upregulated by p53, and a decrease in the mRNA and protein levels of Bcl2, which are downregulated by p53, were observed. The inhibition of Mdm2 by nutlin-3 increased the basal p53 protein level and rescued its tetanization-induced depletion, which suggested the involvement of Mdm2 in the control over p53 during LTP. Furthermore, nutlin-3 caused an increase in the basal expression of Bax and a decrease in the basal expression of Bcl2, whereas tetanization-induced changes in their expression were occluded. These results support the hypothesis that p53 may be involved in transcriptional regulation during the early phase of LTP. We hope that the presented data may aid in the understanding of the contribution of p53 and related genes in the processes that are associated with synaptic plasticity.


Introduction
The storage of information in the brain relies on long-term synaptic plasticity, which depends on complex molecular interactions involving gene expression. One of the forms of synaptic plasticity is hippocampal long-term potentiation (LTP). The late phase of LTP is known to be dependent on mRNA and protein synthesis during a brief time after stimulus [1].
Several transcription factors are rapidly induced in association with LTP [2]. We demonstrated earlier that the tetanization of SC in rat hippocampal slices, which induces the long-term potentiation of CA3-CA1 synapses, is accompanied by a brief (less than 40 min) increase in the binding of transcription factor p53 with the promoter of S100B and by an increase in the level of S100B mRNA [3]. Interestingly, the maximal increase in the DNA-binding activity of p53 coincided with the maximal rate of decrease in the p53 protein level, which suggested the activation of negative feedback to p53.
p53 is a key regulator of the cell cycle and "programmed cell death" (apoptosis). Biological functions of p53 are primarily mediated through the transcriptional regulation of target genes [4,5]. Under stress conditions, the increased activity of a53 can increase the susceptibility of cells to death signals by shifting the balance between proapoptotic (Bax, Noxa, and Puma) and antiapoptotic (Bcl2, Birc5) proteins of the Bcl2 family, which regulate the activity of proteolytic enzymes called caspases. In addition, p53 can induce the expression of "death receptors, " which initiate apoptosis by the binding of their cognate ligands. Thus, upon activation, a53 can induce apoptosis by the activation of caspases through multiple mechanisms [6].
p53 is widely known primarily due to its ability to suppress tumors; however, the list of its functions is growing. Accumulating evidence suggests that p53 should be viewed 2 Neural Plasticity as a crucial decision-maker molecule rather than as a tumor suppressor protein [5,7]. p53, caspases, and Bcl-2 family members can regulate the proliferation and differentiation of neural progenitor cells, as well as neurite outgrowth and regeneration [7,8]. The activation of caspases, which is regulated by Bcl2 family proteins, seems to be necessary for synaptic modifications during long-term depression [9] and to contribute to LTP [10]. Another p53 transcriptional target, microRNA-34a, also regulates neurite outgrowth, spinal morphology, and function [11]. Finally, p53 regulates the transcription of genes that encode secreted proteins, such as interleukin 6, TNF-, and S100B, which can modulate synaptic plasticity [12,13].
Thus, which functional consequences might the brief activation of p53 have? The control of the transcriptional activity of p53 is considered crucial for determining which p53 response is activated [5]. In this paper, we describe the expression of p53-related genes in the rat hippocampal CA1 area in the early phase of long-term potentiation (30 min after tetanization) using real-time PCR. Our set of genes includes at least 50 genes that are transcriptionally regulated (directly or indirectly) by p53, as well as other p53-related genes, which are not known p53 transcriptional targets. The posttetanization time point was chosen from our previous data on the tetanization-induced increase in S100B mRNA, which is maximal at 30 min after tetanization [3,14].
p53 transcriptional targets are regulated by multiple factors. One of the tools for the preliminary assessment of the transcription dependency on p53 is the inhibition of Mdm2 by nutlin-3 [15][16][17]. Mdm2 negatively modulates the transcriptional activity, the stability, and the mRNA translation of p53, and the inhibition of Mdm2 results in the activation of p53 and the increase in the p53 protein level [13,[18][19][20]. Therefore, to estimate the contribution of p53 to the tetanization-induced regulation of the genes studied, we reexecuted our experiments using nutlin-3.
Although not definitive, our results allow hypothesizing that p53 participates in transcriptional regulation during the early phase of LTP. We hope that the presented data may aid in the understanding of the physiological function of p53related genes in the processes that are associated with synaptic plasticity.

Experimental Animals.
Male Wistar rats (7-9 weeks of age) were supplied by the Institute of Cytology and Genetics SB RAS (Novosibirsk, Russia). Animals were acclimated for 1 week and were allowed free access to food and water. All experimental procedures were approved by the Animal Care Committee for the Institute of Molecular Biology and Biophysics SB RAMS and were performed in strict accordance with the National Institutes of Health guidelines.

Hippocampal Slice Preparation and Tetanization.
Animals were decapitated, and the brain was rapidly removed and placed in ice-cold oxygenated (95% O 2 , 5% CO 2 ) artificial cerebrospinal fluid (ACSF): 126 mM NaCl, 4 mM KCl, 1.24 mM NaH 2 PO 4 , 1.3 mM MgSO 4 , 2 mM CaCl 2 , 26 mM NaHCO 3 , and 10 mM D-glucose, pH 7.4. The left hippocampus was dissected for removal and cut into 400 m thick transverse slices, using a chopper. Four consecutive slices from the dorsal region of the hippocampus were transferred to a submerged recording chamber. Slices were perfused at a rate of 2 mL/min with fresh, oxygenated ACSF at room temperature (22-24 ∘ C) for 30 min; then the perfusion rate was reduced to 1.5 mL/min, and the temperature was raised to 32-33 ∘ C. For extracellular recordings, the recording electrode, which was filled with ACSF, was placed in the CA1 pyramidal cell layer. To stimulate Schaffer collaterals, stimulating electrodes, which were filled with ACSF, were placed in the stratum radiatum. The intensity of the stimulation was adjusted to obtain a p-spike amplitude that was ∼50% of the maximal response. P-spikes amplitudes were measured as the vertical distance from the peak of the spike to the line joining the peak positivities on either side. Slopes of field excitatory postsynaptic potentials were measured as the slope of the line joining the peak of the presynaptic volley and the peak positivity of the response.
In the experiments that were intended for the preparation of samples for real-time PCR and Western blot analysis, two of the four slices remained nonstimulated throughout the entire incubation period (4 h 15 min) and were used as a baseline control in the corresponding set of samples. Two other slices were tetanized (4 trains of 1 s, 100 Hz stimulations spaced by 30 s intervals) 30 min before the termination of incubation. Electrodes were placed on the slices 2-5 min before the tetanization and were removed immediately after the stimulation. At the end of the incubation, slices were transferred to ice-cold oxygenated ACSF, and the CA1 field was rapidly cut away from each slice, as described previously [14], and placed into the Allprotect Tissue Reagent (Qiagen). Six relevant slices from three animals were pooled to prepare one mRNA/protein sample. Four independent tetanization/control pairs of samples were prepared for each (nutlin-3/vehicle) incubation condition.

cDNA Synthesis and Real-Time PCR.
To identify the participation of p53 in transcriptional regulation during the early phase of LTP, a Rat p53 Signaling Pathway RT 2 Profiler PCR Array (Qiagen) was used. Total RNA was isolated using an AllPrep DNA/RNA/Protein Mini Kit (Qiagen) according to manufacturer's protocol. The RNA concentration was measured using a Quant-iT Assay Kit and a Qubit instrument (Invitrogen); the RNA quality was evaluated by electrophoresis. cDNA was synthesized from total RNA using RT 2 First Strand cDNA Synthesis Kit (Qiagen). Real-time PCR was performed with a Maxima SYBR Green qPCR Master Mix (Thermo Scientific) and a CFX96 Multicolor Detection System (Bio-Rad) according to manufacturer's Neural Plasticity 3 instructions. The obtained data were analyzed using the Excel-based PCR Array Data Analysis Software (Qiagen). The mean of Ct values of five housekeeping genes (Actb, -actin; B2m, beta-2 microglobulin; Hprt1, hypoxanthine phosphoribosyltransferase 1; Ldha, lactate dehydrogenase A; Rplp1, ribosomal protein, large, P1) was used as the reference according to manufacturer's instructions. The concentration of S100B mRNA was measured as described previously [68].

Preparation of Whole-Cell Extracts.
Whole-cell extracts from frozen slices were prepared using an AllPrep DNA/RNA/Protein Mini Kit (Qiagen) according to the manufacturer's protocol. Protein concentrations were determined using a Quant-iT Assay Kit and a Qubit instrument (Invitrogen).

SDS-PAGE Electrophoresis and Western Blot.
Sixty micrograms of whole-cell proteins per lane was separated using 12% SDS-PAGE and transferred to nitrocellulose membrane. Membranes were stained with Ponceau S to verify the loading and transfer efficiency. Immunodetection was performed using anti-p53 (1 : 500, Santa Cruz Biotechnology), anti-bax (1 : 2000), and anti-bcl2 (1 : 1000, Abcam) or anti-actin (1 : 2000, Sigma-Aldrich) antibodies. -actin was used as a control to ensure the equal loading of samples. The bands were visualized using Visualizer Spray and a Glow ECL Western Blotting Detection System (Millipore).

Data
Analysis. The data are expressed as the mean ± S.E.M. The significance was assessed using Student's -tests with the criterion set at < 0.05. In order to exclude the overestimation of the significance of studied genes mRNA fold changes due to possible inappropriate biases in values of housekeeping genes mRNAs, we performed the additional statistical analysis based on calculations made on the assumption that the expression of the housekeeping genes is constant. mRNA fold changes were considered as significant, if they were significant in both standard and additional tests. Besides, fold changes, which ranged from 0.95 to 1.05, were considered as not significant.

Results
We used real-time PCR analysis to study the expression of 85 genes that are functionally related to p53 in the early phase of LTP in the CA1 area of rat hippocampal slices ( Table 1). As described earlier [3,69], our experimental protocol induces robust and enduring potentiation slightly reducing at 3 h after tetanization, which is characteristic of late-LTP produced in rat CA1 by repeated tetanization [70]. The application of vehicle (DMSO 0.1%) in perfusing milieu from 30 min before to 30 min after tetanization did not influence significantly the basal responses (Figure 1), and the potentiation time course also did not differ significantly from that which has been previously described. Preliminary experiments showed that DMSO had no effect on basal expressions and tetanization-induced mRNA fold changes of genes studied (not presented). A short description of the genes studied is presented in Table 2. A variant of their functional classification is available on the PCR array manufacturer's website (http://www.sabiosciences.com/rt pcr product/HTML/PARN-027Z.html). Four transcripts (LOC367198, Myod1, Serpinb5, and Wt1) were not detected in our samples (Ct > 35) and, therefore, were omitted in Table 1. Besides, Esr1, Lig4, Sfn, and Tnfrsf10b are not presented, since their changes were not significant in all the comparisons.  The tetanization-induced changes in expression profiles of genes that are upregulated (p53-URG) or downregulated by p53 (p53-DRG) did not differ significantly. Specifically, the expression of only two p53-DRGs (11%), Bcl2 and Pttg1, significantly decreased, whereas the expression of seven p53-DRGs (37%) significantly increased after tetanization.
Similarly, the expression of two p53-URGs (6%), Bbc3/Puma and Rprm, also decreased, whereas the expression of 18 p53-URGs (58%) increased after tetanization. Thus, the LTPrelated regulation of p53 transcriptional target genes appears to be rather complex, and the contribution of p53 to this regulation seems to not be crucial, in many instances.   To estimate the possible contribution of p53 to the regulation of tetanization-induced gene expression, we inhibited Mdm2 by nutlin-3. Surprisingly, nutlin-3 attenuated basal mRNA levels of only a few p53-DRGs (Bcl2, Prkca, and Pttg1), whereas the expression of most genes increased (Table 1), which partly might be associated with the involvement of Mdm2 in the regulation of mRNA stability [71].
We assumed that necessary (but not sufficient) criteria for a gene, which is regulated predominantly by p53 after tetanization, are (1) the level of mRNA changes significantly after tetanization under normal conditions; (2) nutlin significantly shifts the basal expression in the same direction as tetanization; and (3) in the presence of nutlin, the effect of tetanization is not significant. We found 16 genes, which conformed to these criteria ( Table 1). Six of them (Bax, Bcl2, Cdkn1a, Gadd45a, Mdm2, and Pten) are known p53 target genes and, therefore, might be considered as promising candidates for more detailed studies of the possible involvement of p53 in the tetanization-induced transcriptional regulation. We chose two of these genes, proapoptotic Bax and antiapoptotic Bcl2, that are up-and downregulated by p53, respectively, to confirm the changes in their expression on the protein level. Indeed, Bax and Bcl2 protein contents followed changes in the mRNA contents ( Figure 2).
Consistent with previous results [3], the induction of LTP was accompanied by a decrease in the protein level of p53 (Figures 2(b) and 2(c)). Nutlin-3 increased the average basal p53 protein level and completely blocked its tetanizationinduced depletion (Figures 2(b) and 2(c)). This effect seemed to be associated not only with the inhibition of the Mdm2dependent degradation of p53 but also with its augmented synthesis due to elevated levels of p53 mRNA (Figure 2(a), Table 1).

Discussion
In spite of the remarkable achievements of many researchers in the identification of LTP-related genes, the extremely complex transcriptional program that is activated in neuroglial networks during LTP remains poorly understood. Our previous results [3] suggest that LTP in the hippocampal CA1 area is accompanied by a transient increase in the transcriptional activity of p53. However, we could find only a few reported examples of p53 target genes that are transcriptionally regulated during the early phase of hippocampal LTP and only in the dentate gyrus [72][73][74][75]. Hippocampal subregions differ one from another in their gene expression profiles [76]. Besides, gene expression differences across the septotemporal (dorsal-ventral) axis of the hippocampus within CA1 were demonstrated [77], and the induction of LTP is differently regulated in dorsal hippocampus versus ventral hippocampus [78]. Therefore, we studied the expression of p53-related genes in the early phase of LTP in the rat dorsal hippocampal CA1 area using real-time PCR analysis. Our set of genes was composed of at least 50 genes that are transcriptionally regulated by p53 (directly or indirectly), as well as other genes that are related to p53-dependent processes (Tables 1 and 2).
To evaluate the contribution of p53 to tetanizationinduced expression profile changes, we inhibited the primary negative regulator of p53 Mdm2 by nutlin-3. Mdm2 negatively modulates (partly independently) the transcriptional activity, the stability, and the mRNA translation of p53, and the inhibition of Mdm2 by nutlin results in both the increase in p53 protein level and the activation of p53-dependent transcription [15]. Therefore, we measured the protein level of p53 to assess the effectiveness of nutlin-3 under our experimental conditions. We observed the significant increase in the protein level of p53 in the presence of nutlin-3, which suggested the effective inhibition of Mdm2 and, therefore, the transcriptional activation of p53. However, because Mdm2 may also influence the transcription of p53 targets through p53independent pathways [20,71], the results that were obtained using nutlin-3 are only circumstantial and are not definitive.
When LTP is induced, the activity of p53 cannot correlate with its total protein amount [3], which is not amazing. The activity of p53 is regulated by numerous mechanisms [5,79], and the induction of LTP is associated with the activation of multiple regulatory cascades [1,2,80], which might influence the portion of the activated p53, its localization, and the recruitment of cofactors.
We observed the upregulation of numerous p53 targets, as well as other p53-related genes, at 30 min after tetanization. However, a percentage of the upregulated genes in the p53-URG group insignificantly exceeded that in the p53-DRG group. Therefore, it is evident, for p53-DRGs and, by analogy, extremely likely for p53-URGs, that p53 is not a single mediator of their LTP-associated regulation. The p53 homologue p73 represents an example of such a complex regulation. Tp73 is one of the genes that are upregulated the most after tetanization in our samples. However, this gene can be upregulated not only by p53 but also by p73 itself and by Egr1 [40]. Nutlin-3 only partly simulated the tetanization-induced upregulation of Tp73 and did not effectively occlude the effect of tetanization (Table 1), which suggests the relatively small involvement of p53 in tetanization-induced upregulation of Tp73.
Egr1 is an immediate early gene, which appears to be critical for memory formation and LTP maintenance [2]. However, in response to stress, Egr1 displays a remarkable functional similarity to p53 and p73 [81][82][83][84]. Egr1 target genes overlap those genes of p53. Egr1, p73, and p53 form a network with positive feedback loops, which respond to stress by the prolonged expression of the p53 family of genes, which results in efficient apoptosis [40]. Our results indicate that the activation of this network is strongly restricted during LTP by Mdm2 activity. Specifically, the induction of LTP was followed by a decrease in the p53 protein level, whereas the p53 mRNA level did not change. However, when Mdm2 was inhibited by nutlin-3 not only was the p53 protein level rescued but also p53 mRNA increased at 30 min after tetanization. Thus, Mdm2 seems to effectively attenuate the activity of the Egr1-p73-p53 network during LTP, which might be explained by the capability of Mdm2 to suppress the transcriptional function of both p53 and p73 [85].

Conclusion
Thus, during the formation of LTP in the hippocampal CA1 area, the increase in the transcriptional activity of p53 seems to occur under tight constraints and leads to the selective transcriptional regulation of target genes. The induction of LTP entails the transcriptional upregulation and/or posttranslational activation of negative regulators of p53 (such as Mdm2), which lead to a rapid cessation of p53 activity. Moreover, in some instances, p53 activity is overridden by other neuron activity-dependent transcriptional regulators. The definite functional roles of p53 and related factors in tetanization-induced processes remain to be elucidated, and now the relevance of these factors to excitotoxicity or to abnormal in vitro conditions cannot be excluded. Nevertheless, it is possible that p53-dependent transcriptional program may be an essential part of synaptic activity-driven adaptive processes. Our results may help to understand the physiological function of p53 pathway in the processes associated with synaptic plasticity. However, to further corroborate the functional significance of p53 in transcriptional regulation during LTP, more detailed studies should be performed using chromatin immunoprecipitation, p53-knockdown or mRNA interference models, promoter constructs, and immunohistochemistry.