**4. Discussion**

The major findings of this study are that TRPC6-mediated ERK1/2 activation regulated LONP1 expression as well as mitochondrial dynamics, which were involved in the invulnerability of DGC to SE (Figure 6).

LONP1 is an inducible ATP-stimulated protease, which plays important roles in cell viability by controlling the maintenance of mitochondrial homeostasis/bioenergetics and DNA integrity [7,23–27]. Therefore, LONP1 expression is up-regulated under some pathological conditions such as hypoxia, oxidative stress, and tumorigenesis [6–9]. However, the underlying mechanisms of regulation of LONP1 expression remain incompletely understood. TRPC6 modulates cell proliferation, differentiation and neuronal vulnerability to various insults [10–12,28,29]. In addition, TRPC6 activates ERK1/2 [15,30], which is involved in mitochondrial dynamics and LONP1 expression [9,31–36]. In the present study, we found that TRPC6 siRNA effectively reduced ERK1/2 activity (phosphorylation) and LONP1 expression under physiological conditions. In contrast, hyperforin, a TRPC6 activator [18,37], increased ERK1/2 activity and LONP1 expression, which were abrogated by U0126 co-treatment. Since LONP1 siRNA did not affect TRPC6 expression and ERK1/2 phosphorylation in the present study, our findings indicate that, at least in DGC, TRPC6-ERK1/2 signaling pathway is one of the up-stream regulators of LONP1 expression.

**Figure 6.** Scheme of roles of TRPC6 in LONP1 expression and mitochondrial dynamics based on the present data and previous reports [11,15,16,20,30]. TRPC6 activation increases Ca2+ influx in DGC. Intracellular Ca2+ activates calcineurin and ERK1/2. Activated calcineurin inhibits the NMDA receptor. In addition, ERK1/2 activation up-regulates LONP1 expression and DRP1 phosphorylation at the 616 site. Subsequently, phosphorylated DRP1 facilitates mitochondrial fission. Thus, TRPC6 may be involved in the quality controls of mitochondria as well as mitochondrial dynamics, which would enhance DGC invulnerability to SE.

LONP1 is required for the maintenance and expression of the mitochondrial enzymes and genomes [25–27]. In particular, LONP1 plays a direct role in the turnover of cytochrome c oxidase (COX), which is a terminal enzyme of the mitochondrial electron transport chain [38–40]. Under a hypoxic condition, LONP1 degrades isoform 1 of COX subunit 4 (COX4-1) to facilitate the switch from COX4-1 to COX4-2 for enhancing mitochondrial respiration [41]. LONP1 also removes the impaired human mitochondrial transcription factor A (TFAM) that is essential for mitochondrial DNA synthesis and its packaging [42–44]. Thus, deregulation of LONP1 leads to cell death by loss of mitochondrial functions [27,45,46]. In the present study, TRPC6 siRNA, LONP1 siRNA, and U0126 exacerbated SE-induced DGC degeneration. In addition, co-treatment U0126 abrogated the protective effect of hyperforin on DGC damage against SE. Therefore, our findings sugges<sup>t</sup> that the TRPC6-ERK1/2 signaling pathway may play a neuroprotective role against SE by regulating LONP1-mediated mitochondrial homeostasis/bioenergetics. Further studies are needed to elucidate the specific targets controlled by LONP1, which would be involved in SE-induced neuronal death.

On the other hand, ERK1/2 activation accelerates mitochondrial fission via dynamin-related proteins 1 (DRP1)-serine (S) 616 phosphorylation [35,36]. Indeed, the blockade of TRPC6 functionality results in aberrant mitochondrial elongation by abrogating ERK1/2-mediated DRP1 activity in DGC [14,15]. Mitochondria are dynamic organelles responsible for generating ATP. In addition, mitochondrial dynamics participate in the synthesis of reactive oxygen species (ROS). Aberrant mitochondrial elongation inhibits mitochondrial respiratory function that triggers excessive ROS production. Excessive mitochondrial fission also impairs the detoxification of excess ROS and extrusion of intracellular Ca2+ [47,48]. Thus, imbalance of mitochondrial fission-fusion induces balance results in neuronal necrosis or apoptosis following SE [15,17,49–53]. Under physiological conditions, furthermore, mitochondrial fission directly enables increases mitochondrial ROS production [1,2]. Considering the relevance between mitochondrial dynamics and ROS syntheses, it is likely that the clearance of oxidized and misfolded proteins generated by ROS may be essential for cell viability. In the present study, TRPC6-mediated ERK1/2 activation facilitated mitochondrial fission, accompanied by LONP1 over-expression. However, LONP1 siRNA resulted in a massive DGC degeneration that was greater than the levels caused by TRPC6 siRNA and U0126, although it did not affect mitochondrial length. Unlike mitochondrial dynamics-related molecules (such as DRP1, optic atrophy 1, and mitofusin 2), LONP1 is up-regulated in response to harmful stresses [54]. Furthermore, LONP1 knockdown does not influence the activities of DRP1 and ERK1/2 under physiological- and post-SE conditions [17]. Since deregulation of LONP1 leads to cell death [17,27,54], the present data indicate that LONP1 may act as one of the important housekeeping antioxidants in mitochondria by limiting oxidative damage

to tolerable levels, regardless of aberrant mitochondrial dynamics. Therefore, our findings sugges<sup>t</sup> that TRPC6-ERK1/2-mediated LONP1 regulation may take part in the quality controls of mitochondria via degradation of oxidized/damaged proteins [23–25] and maintenance of mitochondrial DNA levels [27] during mitochondrial fission under physiological- and pathological conditions.

In the present study, TRPC6 siRNA increased seizure susceptibility in response to pilocarpine. TRPC6 inhibits N-methyl-d-aspartate (NMDA) receptor activity mediated by calcineurin [11]. Indeed, TRPC6 knockdown increases the excitability ratio (an index of synaptic efficacy, also referred as excitatory postsynaptic potential-population spike amplitude coupling) [14,16] indicating the lowering intrinsic threshold of neuronal firing in postsynaptic neurons [55]. Therefore, TRPC6 knockdown reduces seizure threshold of DGC via the heightened efficacy of NMDA receptor function in DGC itself [16]. Furthermore, TRPC6 siRNA reduces γ-aminobutyric acid (GABA)-ergic inhibitions onto the DGC during and after high-frequency stimuli due to the impaired repetitive firing of interneurons [16]. However, the present data show that TRPC6 activation by hyperforin did not affect seizure susceptibility in response to pilocarpine. Unlike TRPC6 knockdown, hyperforin shows the distinct effects on evoked potentials in a dose-dependent manner. The higher concentrations of hyperforin (10 and 100 μM) reduce the population spike amplitude (an indicative of synchronous postsynaptic discharges [14,16]), while a lower concentration (1 μM) increases it [56]. Consistent with the present data, the concentration of hyperforin (6 μM) cannot affect GABAergic inhibition and the seizure susceptibility in response to pilocarpine due to the functional saturation of Kv4.3 channels in interneurons, unlike DGC [16,18]. Furthermore, Sell et al. [57] have reported that hyperforin induces TRPC6-independent H<sup>+</sup> currents in HEK-293 cells, cortical microglia, chromaffin cells, and lipid bilayers. This action of hyperforin as a protonophore leads to cytosolic acidification and subsequently increases free intracellular Na<sup>+</sup> concentration via Na<sup>+</sup>-H<sup>+</sup> exchanger (NHE). Thus, it is plausible that this unspecific properties of hyperforin as protonophore may be also involved in the ineffectiveness of hyperforin on pilocarpine-induced seizure activity. This is because seizure activity results in biphasic pH shifts, consisting of an initial extracellular alkalinization, followed by a slower acidification. The early extracellular alkalosis increases excitability because of reductions in GABAA receptor inhibition and enhancement in NMDA receptor currents, and the extracellular acidosis is involved in seizure termination [58]. Thus, it is presumable that hyperforin-induced H<sup>+</sup> efflux from neurons or glia would attenuate seizure activity in response to pilocarpine, independent of TRPC6. However, the simultaneous Na<sup>+</sup> accumulation would offset the inhibitory effect of extracellular acidosis on neuronal excitability by causing a lowering of the threshold for action potential generation in neurons and reducing the driving force for Na<sup>+</sup>-dependent re-uptake of glutamate and other excitatory neurotransmitters into glia or neurons [57,59–61]. Thus, it is likely that these discrepancies of hyperforin from TRPC6 siRNA may lead to the ineffectiveness of hyperforin on seizure susceptibility to pilocarpine in the present study.
