miR‐485's anti‐drug resistant epilepsy effects by regulating SV2A/PSD‐95 and targeting ABCC1 and neuronal signaling‐transduction proteins in hippocampus of rats

Abstract Aim Drug‐resistant epilepsy (DRE), most subsequently developing refractory epilepsy, causes a significant burden to the society. microRNAs have been demonstrated as key regulators and therapeutic targets in epilepsy. Accordingly, the aim of the present study was to test whether miR‐485 could be a potential target for DRE. Methods and results An in vivo DRE model was developed in Sprague–Dawley rats by lithium chloride‐pilocarpine and screened by antiepileptic drugs. We found that miR‐485‐5p in hippocampus was significant downregulated at early stage and recovered to normal level at late stage of DRE. Overexpression of miR‐485‐5p in dentate gyrus (DG) of hippocampus in DRE rats could significantly decrease the frequency of seizures and the numbers of epileptiform spikes of hippocampal DG neuron, and could specifically decrease SV2A expression without affecting PSD‐95 expression in DG. Furthermore, miR‐485‐5p overexpression could significantly downregulate the expression of efflux transporter related to multidrug resistance (ABCC1) in hippocampus at late stage of DRE. Finally, a specific expression pattern of neuronal signaling‐transduction proteins (LRP4, MDM4, p53, and TMBIM1) for DRE was observed, and miR‐485‐5p overexpression could modulate these proteins’ expression levels toward normal in hippocampus both at early and late stage of DRE. Conclusion Collectively, these results suggest that miR‐485 was a potential target for anti‐DRE, and this effects might be partially via miR‐485‐5p/homeostatic‐synaptic plasticity‐molecule axis and/or targeting efflux transporter (ABCC1) and other neuronal signaling‐transduction proteins (LRP4, MDM4, p53, and TMBIM1).

few decades, the therapeutic efficacy is still limited (Tang et al., 2017).
Most DRE patients subsequently develop refractory epilepsy (Kwan & Brodie, 2000;Kwan et al., 2010) and suffer from severe psychosocial problems and even increased morbidity and mortality (French, 2007).
It has been proposed that refractory epilepsy could be mediated by several mechanisms relating to seizure severity, pharmacokinetic, gene variant, neural network, and the transporter such as the efflux transporter correlates with pharmacoresistance (Potschka, 2012;Tang et al., 2017).
Recently, microRNAs (miRNAs), a class of endogenous short noncoding RNAs that modulate gene expression at the posttranscriptional level, inducing the degradation or inhibition of the translation of target genes (Bartel, 2009;Gisel et al., 2014;Shukla et al., 2011), have received increasing attention. Results of previous studies suggested a possible role of miRNAs in epilepsy's pathophysiology since epileptogenic and molecular profiles found a number of miRNAs changed in epileptic hippocampus of both animal models and human tissues, and it was also proposed as therapeutic targets for epilepsy. Furthermore, recent study results have indicated a role of miRNAs in reversing pharmacoresistance in the CNS of epilepsy (Liu et al., 2012;. miR-485 is one of miRNAs that significantly downregulated in hippocampus of epilepsy (Gorter et al., 2014), and miR-485 is critical in controlling homeostatic synaptic plasticity (Cohen et al., 2011). Homeostatic synaptic plasticity is also one of the key epileptogenesis mechanisms, and we tested the anti-DRE effects of miR-485 and whether these effects were via regulating homeostatic synaptic plasticity molecules and/or targeting efflux transporter and other neuronal signaling-transduction proteins.

Ethical statement
The Sprague-Dawley (SD) rats were bought from the Zhejiang University Animal Center. All procedures were approved by the Zhejiang University Animal Experiment Ethics Committee and followed the Zhejiang University Guidelines on Animal Care.

Western blots
Rat hippocampus was lysated in RIPA Lysis Buffer (

Histological preparation of tissue and immunostaining
Rats were terminated at 14 weeks after SE under deep isoflurane anesthesia and decapitated for rapid brain dissection on ice. After fixed in 3% paraformaldehyde overnight at 4 • C and then dehydrated in cacodlyate-PBS containing 15−30% sucrose at 4 • C, brains were processed for cryo-sectioning at 40 µm thickness (coronal/sagittal) on microtome. Free-floating brain sections were mounted on glass slides and imaged for AAV-GFP or stored in glycerol at −20 • C for immunofluorescent staining.

Statistical analyses
All results are expressed as the mean ± SD. Statistical results were analyzed using SPSS 19.0 software (SPSS, USA). Differences between groups in the levels of Western blot and qPCR were analyzed using unpaired Student's t-test. A value of p < .05 was considered statistically significant.

The dynamic expressions of miR-485 in hippocampus of DRE rats
miR-485 is one of the significantly downregulated miRNAs in hippocampus of epilepsy and is critical in controlling homeostatic synaptic plasticity in the brain. We asked whether the miR-485-5p was changed in hippocampus of DRE rats. We found that the mRNA level was significantly downregultaed in DRE rats at 6 weeks (n = 6) and 7 weeks (n = 3) after SE compared to the control (6 weeks and 7 weeks, n = 3) and epilepsy (6 weeks and 7 weeks, n = 3) groups, respectively (p < .05, Figure 1), but the level recovered to the normal level at 9 weeks (n = 3) after SE, which was not significantly different from that of the control (n = 3) and epilepsy (n = 3) groups, respectively (p > .05, Figure 1).

F I G U R E 2
The AAV effectively infected the hippocampus as indicated by immunofluorescence microscopy (200×) F I G U R E 3 miR-485-5p overexpression decreased the epileptic-discharges of DRE rats. (a) The numbers of seizures (grade ≥ 4) in 12 h/d*7 were significantly decreased by the AAV-miR-485-5p infection at 14 weeks after SE (n = 4, p < .05). (b and c) miR-485-5p overexpression decreased the epileptic-discharges of hippocampal neurons of rats with DRE. (b) Samples of whole cell current recording using hippocampal slices of DRE rats. (c) Barographs of the numbers of epileptiform spikes in hippocampal slices of DRE rats (15 neurons were recorded from three rats for each group). * p < .05

miR-485-5p overexpression in hippocampus decreased epileptic-discharges of DRE rats
To further determine whether miR-485-5p could inhibit the epilepticdischarges of DRE rats, AAV-miR-485-5p was injected into DG of DRE rats using a stereotaxic frame. Four weeks after injection, the AAV effectively infected the brain tissue as indicated by immunofluorescence microscopy (Figure 2). Saline and AAV-GFP served as controls. At 14 weeks after SE, AAV-miR-485-5p-infected DRE rats were subjected to seizure evaluation using video-recording for 12 h/d*7 and whole-cell current recording using hippocampal slices.
The numbers of epileptiform spikes were also significantly reduced by the AAV-miR-485-5p infection of DG neurons (p < .01, 15 neurons were recorded from three rats for each group) (Figure 3b and c).

miR-485-5p overexpression affected expressions of homeostatic synaptic plasticity-related proteins in hippocampus of DRE rats
miR-485 is critical in controlling homeostatic synaptic plasticity (Cohen et al., 2011) via targeting SV2A to regulate dendritic spine density

miR-485-5p overexpression affected expressions of multidrug resistance-associated protein (MRP1/ABCC1) and neuronal signaling-transduction proteins in hippocampus of DRE rats
Multidrug resistance due to efflux transporters has been studied extensively. ABCC1 is one of the best understood efflux transporters.
Compared with the control group, at 6 weeks after SE, the expression levels of ABCC1 in epilepsy group were not significantly increased (p > .05, Figure 5a and b), but significantly increased in the DRE group (p < .05 Figure 5a and b); at 7 weeks after SE, the levels in epilepsy group were similar to that of the control group (p > .05 Figure 5a and b), but significantly decreased in the DRE group (p < .05 Figure 5a and b); at 9 weeks after SE, the levels in both epilepsy and DRE groups were significantly increased (p < .05 Figure 5a and b).
We reviewed literatures and selected four cell signalingtransduction proteins (LRP4, MDM4, p53, and TMBIM1) to check their changes in hippocampus of DRE rats. At 6 weeks after SE, compared with the control group, the expression levels of MDM4, p53, and TMBIM1 were significantly increased in the epilepsy group (p < .05, Figure 6a and b), while the levels of LRP4 and MDM4 were significantly decreased (p < .05, Figure 6a and b) but the levels of p53

F I G U R E 5
The expressions of drug-resistant-related proteins (ABCC1) in hippocampus of DRE rats at 6, 7, and 9 weeks after SE. *p < .05 versus the control group; # p < .5 versus the epilepsy group were significantly increased in the DRE group (p < .05, Figure 6a and b), and the levels of LRP4, MDM4 and TMBIM1 were significantly downregulated in the DRE group compared with the epilepsy group (p < .01, Figure 6 a and b). At 7 weeks after SE, compared with control group, the levels of LRP4 and MDM4 were significantly increased in epilepsy group (p < .05, Figure 6a and b), while only MDM4 was significantly increased in the DRE group (p < .05, Figure 6a and b), and the levels of LRP4, MDM4, p53, and TMBIM1 had no significant changes in the DRE group compared with the epilepsy group (p > .05, Figure 6a and b). At 9 weeks after SE, compared with the control group, the levels of MDM4 and p53 were significantly increased in the epilepsy group (p < .05, Figure 6a and b); the levels of MDM4 were significantly increased in the DRE group (p < .05, Figure 6a and b), and the levels of LRP4, MDM4, p53, and TMBIM1 in the DRE group had no significant difference compared with the epilepsy group (p > .05, Figure 6a and b).
We then tested whether miR-485-5p overexpression affected expressions of ABCC1 at 9 weeks after SE and expressions of cell signaling-transduction proteins (LRP4 and p53 at 6 weeks after SE, and MDM4 at 9 weeks after SE) in hippocampus of DRE rats. We found miR-485-5p overexpression could significantly modulate the levels of ABCC1, LRP4, MDM4, and p53 toward normal in hippocampus of DRE rats (p < .05 Figure 7a-d).

DISCUSSION
Most DRE patients subsequently develop refractory epilepsy (Kwan & Brodie, 2000;Kwan et al., 2010) and suffer from severe psychosocial problems (French, 2007) due to therapy being still limited. It would be of a great significance to find a novel target for DRE and its relevant Furthermore, studies have demonstrated that miRNAs could directly reverse pharmacoresistance in the epileptic CNS (Liu et al., 2012;. miR-485 is one of miRNAs that significantly downregulated in epileptic hippocampus (Gorter et al., 2014), and is critical in controlling homeostatic synaptic plasticity (Cohen et al., 2011).
Homeostatic synaptic plasticity is also one of the key mechanisms of epileptogenesis; herein we checked the miR-485′s anti-DRE effects and whether these effects were via regulating homeostatic synaptic plasticity.
In the present study, we found that miR-485-5p in hippocampus of DRE was significantly downregulated at early stage and recovered toward normal at late stage, while its levels were not significantly changed in the epilepsy group. Our results showed that in the epilepsy group, miR-485-5p was not significantly changed in hippocampus. This result is different from the previous reports (Gorter et al., 2014), which might be due to different methods of model preparation, observation time points, etc. Then we tested whether miR-485-5p overexpression at early stage could inhibit the genesis of DRE? We found that miR-485-5p overexpression in DG of DRE rats did reduce the numbers of epileptiform spikes (p < .01) (Figure 7a and b). Our results indicated that miR-

485-5p was a potential targets for anti-DRE.
Previous study has demonstrated that miR-485 was critical in controlling homeostatic synaptic plasticity (Cohen et al., 2011) via targeting SV2A to regulate dendritic spine density and PSD-95 clustering.
Homeostatic synaptic plasticity is also one of the key epileptogenesis mechanisms; we then determined the effects of miR-485-5p overexpression on the expressions of PSD95 and SV2A in hippocampus of DRE rats. We found that miR-485-5p overexpression (AAV-miR-485-5p infection) specifically decreased SV2A expression without affecting  (Cohen et al., 2011). It has been demonstrated that SV2A knockdown could mimic the effects of miR-485 overexpression, and inhibiting endogenous miR-485 could block the homeostatic reduction in spine density, PSD-95 puncta density, and GluR2 expression (Cohen et al., 2011). As these morphological changes were significantly reduced after increasing synaptic activity, it was supposed that the effects of miR-485 on synapse-associated transcripts, including SV2A, are consistent with a possible role of this miRNA in homeostatic synaptic plasticity acting to reduce transcripts and/or suppress translation to decrease synaptic connectivity under conditions of persistent hyperactivity such as DRE.
The transporter mechanism is the most widely accepted and investigated theory for refractory epilepsy. It is postulated that transporters actively transport substrates, including AEDs, against their concentration gradient out of neurons and blood-brain barrier, limiting their entry into the neurons or brain, and thereby causing resistance (Sisodiya et al., 2006). Previous study found that transporter mRNA was overexpressed in brain tissue resected from DRE patients. Therefore, this hypothesis is based on that overexpression of efflux transporter correlates with pharmacoresistance in DRE, and ASDs are subject to active transport by efflux transporters (Sisodiya et al., 2006 LRP4 plays crucial roles in the adult CNS, including formation of neuromuscular junctions in the peripheral nervous system, synaptogenesis in the developing brain, maintenance of excitatory synaptic transmission, hippocampal synaptic plasticity, fear conditioning, associative and spatial learning, and long-term potentiation (Gomez et al., 2014;Pohlkamp et al., 2015;Sun et al., 2016;Karakatsani et al., 2017).
Moreover, LRP4 protein has been detected in postsynaptic membrane fractions purified from adult rat forebrain where it interacts with the postsynaptic scaffold protein PSD95 (Gomez et al., 2014;Tian et al., 2006