METH-induced behavioral sensitization
The model of behavioral sensitization was established by 7 days of METH (5.0 mg/kg, s.c.) administration followed by 7 days of withdrawal and challenged with a single METH injection (1.0 mg/kg, s.c.) on day 15. One-way ANOVA showed a difference between the groups (F(2, 21) = 14.58, P<0.001). Rats in the METH sensitization group (METH-METH group) showed significantly greater locomotor activity than either the saline control (NS-NS) or METH acute treatment group (NS-METH) (Tukey’s post hoc test, P<0.001 compared with the NS-NS group; P<0.01 compared with the NS-METH group, Figure 1B), indicating the increased locomotor response of sensitization. Compared with the NS-NS group, rats in the NS-METH group showed longer distances (P<0.05, Figure 1B).
Changes in mRNA expression in the PFC of METH-induced behavioral sensitization rat
An mRNA microarray was performed to determine a genome-wide profile of METH-induced behavioral sensitization. The cut-off value was set at a fold change of ≤0.5 or ≥2.0 (P<0.01). Figure 2 shows the clustering diagram and a heat map of the changes in mRNA expression in the NS-NS, NS-METH, and METH-METH groups. The expression of a large number of genes was altered by METH exposure (P<0.01) (Figure 2B). Among these, 275 genes were differentially expressed in the METH-METH group compared with the NS-METH group (97 genes were upregulated and 178 genes were downregulated), and 232 genes displayed differential expression in the NS-METH group compared with the NS-NS group (122 genes were upregulated and 110 genes were downregulated) (Table S2).
Ontology and pathway analyses of differentially expressed mRNAs in the PFC of METH-induced sensitization rats
We performed a GO analysis and categorized the differentially expressed genes to investigate whether the clustering of differentially expressed genes induced by METH exposure correlates with functional categories. Figure 3A and Table S3 show the top 10 ranked GO terms and the number of genes in each category. Compared with the acute METH treatment (NS-METH), the biological processes that were markedly increased in the METH-induced behavioral sensitization (METH-METH) group included biological regulation, cellular process and organism process. Regarding the molecular function, binding and protein binding were enriched in the METH-METH group.
We further performed a KEGG pathway enrichment analysis to assess the functional features of METH-induced behavioral sensitization-mediated alterations in gene sets. The top 10 significantly altered pathways are shown in Figure 3B and Table S4. Many signal transduction pathways were enriched in the METH-induced behavioral sensitization group, including cell adhesion, PI3K-AKT signaling pathway, olfactory transduction, and cell cycle (METH-METH group vs. NS-METH group, P<0.0001). For the METH acute treatment (NS-METH) group, the genes categorized by the KEGG analysis were enriched in the cell adhesion, the pathway in cancer, and dopaminergic pathway.
Changes in histone acetylation in the PFC of METH-induced sensitization rats
ChIP coupled with a DNA microarray analysis revealed that METH increased histone acetylation (H3 or H4) on a large number of gene promoters (Figure 4), in accordance with mRNA activation. Compared with the NS-METH group, the METH-METH group induced many more acetylation modifications of H3 than H4 on the promoters; specifically, 821 genes presented H3 hyperacetylation and 10 genes showed H4 hyperacetylation (Table S5). The NS-METH treatment also induced hyperacetylation on the gene promoters, including 947 genes with H3 hyperacetylation and 4902 genes with H4 hyperacetylation. Very little H3/H4 hyperacetylation overlapped between the METH-METH and NS-METH groups.
Real-time PCR confirmation of the candidate genes
Based on the mRNA microarray data, GO/KEGG enrichment analyses, and the analysis of the functional properties, the genes listed in Table S6 were likely to be associated with METH-induced behavioral sensitization. Of these genes, Avp, Bcl2l1, Egr1, E2f3, Lnx2, Shoc2, Stx2 and Zfp36 clustered into the top 10 GO and KEGG categories, and Anp32a, Eml2, Exog, Hira, Metrn, Pou3f2, Stk32, Syt8, and Trim17 were included in the GO category. According to the analysis of the GO categories, the expression of Lnx2, Shoc2 and Stx2 was altered by a single METH injection, and the expression of the other genes listed above was affected by METH-induced behavioral sensitization (METH-METH) but not the acute METH treatment (NS-METH). Furthermore, all of the genes displayed H3/H4 hyperacetylation. We then used qPCR to confirm the expression of some genes of interest. METH challenge in the behavioral sensitization model caused a significant increase in the mRNA levels of acidic nuclear phosphoprotein 32 family member A (Anp32a), calcium/calmodulin-dependent protein kinase II inhibitor 1 (Camk2n1), echinoderm microtubule-associated protein-like 2 (Eml2) and POU class 3 homeobox 2 (Pou3f2), and a decrease in syntaxin 2 (Stx2), tripartite motif-containing 17 (Trim17), and zinc finger protein 36 (Zfp36) (P<0.05, one-way ANOVA followed by Turkey’s post hoc test, Figures 5 and 6).
Alterations in ANP32A and POU3F2 expression in the development, withdrawal and challenge periods of METH-induced sensitization
Then, we selected ANP32A and POU3F2 and further measured the mRNA expression and histone modification of these two genes. In addition to the challenge of behavioral sensitization, the development and withdrawal phases were also involved to investigate the changes in ANP32A and POU3F2 expression throughout the whole process of METH-induced behavioral sensitization. After a chronic METH (5 mg/kg, s.c.) treatment for 7 days (development phase), the levels of the ANP32A mRNA and H4 acetylation were markedly increased compared with the NS group (P<0.05, t-test, Figure 6A-6B). However, the expression of the ANP32A mRNA returned to the normal level, while H4 acetylation remained at a high level after 7 days of withdrawal (P<0.05, t-test). Then, an injection of METH (1 mg/kg, s.c.) was administered as a challenge on day 15. One-way ANOVA showed a significant difference in mRNA expression (F(2, 18) = 8.824, P<0.01) and H4 acetylation (F(2, 26) = 6.072, P<0.01) between groups, and Tukey’s post hoc test showed that the challenge induced the expression of ANP32A mRNA (P<0.01, METH-METH vs. NS-NS) and H4 acetylation (P<0.01, METH-METH vs. NS-METH, P<0.05, METH-METH vs. NS-NS) in the METH pretreatment group (METH-METH). The injection of METH alone also increased the level of the ANP32A mRNA (P<0.05, NS-METH vs. NS-NS), but did not affect histone acetylation.
We next measured the change in levels of the POU3F2 mRNA and histone modification. The expression of the POU3F2 mRNA was significantly increased by the chronic METH treatment for 7 days and decreased after 7 days of withdrawal (P<0.05, t-test), while H3 and H4 acetylation were not affected by the treatment (Figure 6C-6D). In the challenge phase, one-way ANOVA showed that challenge with a low dose of METH caused a significant difference in mRNA expression (F(2, 45) = 4.01, P<0.05) and H3 and H4 hyperacetylation (H3: F(2, 36) = 4.17, P<0.05; H4: F(2, 38) = 3.47, P<0.05) between groups. Tukey’s post hoc test revealed a significant increase in POU3F2 mRNA expression (P<0.05, METH-METH vs. NS-NS) and H3 and H4 hyperacetylation in the behavioral sensitization group (H3: P<0.05, METH-METH vs. NS-NS, METH-METH vs. NS-METH; H4: P<0.05, METH-METH vs. NS-NS). Notably, the single METH injection (NS-METH) also increased the expression of the POU3F2 mRNA (P<0.05, vs. NS-NS), but not the levels of H3/H4 acetylation.