Rapid regulation of microRNA following induction of long-term potentiation in vivo

Abstract

Coordinated regulation of gene expression is essential for consolidation of the memory mechanism, long-term potentiation (LTP). Triggering of LTP by N-methyl-D-aspartate receptor (NMDAR) activation rapidly activates constitutive and inducible transcription factors, which promote expression of genes responsible for LTP maintenance. As microRNA (miRNA) coordinate expression of genes related through seed sites, we hypothesize that miRNA contribute to the regulation of the LTP-induced gene response. MiRNA function primarily as negative regulators of gene expression. As LTP induction promotes a generalized rapid up-regulation of gene expression, we predicted a complementary rapid down-regulation of miRNA levels. Accordingly, we carried out global miRNA expression profiling in the rat dentate gyrus 20 min post-LTP induction in vivo. Consistent with our hypothesis, we found a large number of differentially expressed miRNA, the majority down-regulated. Detailed analysis of miR-34a-5p and miR-132-3p revealed this down-regulation was transient and NMDAR-dependent, whereby block of NMDARs released an activity-associated inhibitory mechanism. Furthermore, down-regulation of mature miR-34a-5p and miR-132-3p occurred solely by post-transcriptional mechanisms, occurring despite an associated up-regulation of the pri-miR-132 transcript. To understand how down-regulation of miR-34a-5p and miR-132-3p intersects with the molecular events occurring following LTP, we used bioinformatics to identify potential targets. Previously validated targets included the key LTP-regulated genes Arc and glutamate receptor subunits. Predicted targets included the LTP-linked kinase, Mapk1, and neuropil-associated transcripts Hn1 and Klhl11, which were validated using luciferase reporter assays. Furthermore, we found that the level of p42-Mapk1, the protein encoded by the Mapk1 transcript, was up-regulated following LTP. Together, these data support the interpretation that miRNA, in particular miR-34a-5p and miR-132-3p, make a surprisingly rapid contribution to synaptic plasticity via dis-inhibition of translation of key plasticity-related molecules.

Keywords: long-term potentiation; maintenance; memory; microRNA; synaptic plasticity.

FIGURE 1

Induction of robust long-term potentiation (LTP) in the perforant path of awake adult rats. (A,B) Average (±SEM) field evoked postsynaptic potential (fEPSP) and population spike (PS) responses expressed as percentage of baseline values. At least a 10% increase in the fEPSP and 50% increase in the PS was required for acceptance as LTP induction. Delta-burst high-frequency stimulation (DBS) elicited by 50T resulted in a reliable and robust LTP (filled circles) that was blocked by the N-methyl-D-aspartate receptor (NMDAR) antagonist CPP (open circles). Inset waveforms are averages of 10 sweeps taken just before DBS and 15–20 min after. Calibration bars: 5 ms; 5 mV. (C) DBS induced a significant increase in Arc mRNA levels, a well-characterized LTP-regulated gene that was also blocked by CPP. RT-qPCR was performed in triplicate and Arc expression normalized to HPRT. Expression values: average ± SEM; one sample two-tailed t-test, *p < 0.05.

FIGURE 2

Long-term potentiation rapidly regulates miRNA 20 min post-LTP. (A) Rapid regulation of miRNA levels revealed by Affymetrix miRNA array analysis. (B) Sixty-five miRNA were found to be differentially expressed using dual selection criteria (two-tailed Student’s t-test p < 0.05; fold change ± 0.15), with 17 up-regulated and 48 down-regulated.

FIGURE 3

Long-term potentiation differentially regulates mature miR-34a-5p and miR-132-3p. (A) Down-regulation of miR-34a-5p was observed 20 min post-LTP, (microarray data validated using RT-qPCR) that persisted for 5 h (Ryan et al., 2012) but returned to baseline by 24 h and was blocked by CPP (20 min; B) Down-regulation of miR-132-3p at 20 min (but not 5 or 24 h post-LTP) was blocked by CPP. RT-qPCR was performed in triplicate and expression normalized to Y1 RNA. Expression values: average ± SEM; one sample two-tailed t-test, *p < 0.05; independent two-tailed t-test, #p < 0.05.

FIGURE 4

Long-term potentiation differentially regulates the primary miRNA transcripts of miR-34a-5p and miR-132-3p. (A) Pri-miR-34a was not differentially expressed 20 min, 5 or 24 h post DBS, or sensitive to CPP. (B) miR-132/212 expression was significantly increased at 20 min, unchanged at 5 h and significantly decreased at 24 h. This effect was partially blocked by CPP. Primary transcripts were normalized to HPRT. Expression values: average ± SEM; one sample two-tailed t-test *p < 0.05.

FIGURE 5

MiR-132-3p binds to wild-type and mutant MRE of target gene transcripts. Interactions between miR-132-3p and (A) four wild-type target MREs, and positive control and (B) miR-132-3p and three mutant target MREs, and positive control, as determined by dual luciferase assays. Significant down-regulation of the luminescence ratio with the miR-132-3p mimic when compared to the negative control was observed for wild-type Hn1, Klhl11, and Mapk1. Gria2 showed no significant regulation. With mutant sequences, significant down-regulation of the luminescence ratio with the miR-132-3p mimic when compared to the negative control was observed for mutant Hn1, and Klhl11. Mutant Mapk1 showed no significant regulation. Normalized to ratio of plasmid only. Average ratio ± SD, n = 4, independent two-tailed t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 6

Mapk1 protein levels are dynamically altered in response to LTP induction. Western blot analysis showed a modest reduction in p42-Mapk1 levels at 20 min, which returned to baseline by 5 h and carried on to become an increase by 24 h post LTP induction. P42-Mapk3 levels were down-regulated at 20 min only. Data were normalized to tubulin and expressed relative to the matched contralateral control hemisphere (average ratio ± SEM, two-tailed t-test, *p < 0.05, **p < 0.01).