**3. Results**

#### *3.1. CDDO-Me Distinctly A*ff*ects SE-Induced Neuronal Death in the Hippocampus*

Figure 1 shows that SE resulted in up-regulation of LONP1 to ~1.5-fold of the control level in the whole hippocampus, accompanied by massive neuronal death in the CA1 region and the hilus of the dentate gyrus (*p* < 0.05 vs. control animals, respectively; Figure 1A,B,E,F). CDDO-Me did not a ffect the seizure latency and its severity in response to pilocarpine (Figure 1C,D). Consistent with a previous report [22], CDDO-Me did not a ffect LONP1 protein levels in the whole hippocampus of control- and post-SE animals, as compared to vehicle (Figure 1A,B). CDDO-Me e ffectively attenuated SE-induced neuronal loss in the CA1 region, but not in the hilus region (*p* < 0.05 vs. vehicle; Figure 1E–F), although it abolished 4-hydroxy-2-nonenal (4-HNE, the end-product of lipid peroxidation) signals in both regions (*p* < 0.05 vs. vehicle; Figure 1E,G). These findings indicate that CDDO-Me may di fferently affect the regional specific neuronal death following SE, independent of ROS generation.

**Figure 1.** Effects of 2-Cyano-3,12-dioxo-oleana-1,9(11)-dien-28-oic acid methyl ester (CDDO-Me) on Lon protease-1 (LONP1) expression, seizure activity and neuronal damage in response to pilocarpine. (**A**–**B**) Effect of CDDO-Me on LONP1 expression in response to pilocarpine. Pilocarpine-induced status epilepticus (SE) increases LONP1 expression, which is not affected by CDDO-Me. (**A**) Representative western blots of LONP1 expression. (**B**) Quantification of LONP1 expression based on western blot data. Open circles indicate each individual value. Horizontal bars indicate mean value (mean ± S.E.M.; \* *p* < 0.05 vs. control animals, respectively; *n* = 7). (**C**–**D**) Effect of CDDO-Me on seizure activity in response to pilocarpine. CDDO-Me does not influence seizure activity induced by pilocarpine. (**C**) Representative electroencephalogram (EEG) traces and frequency-power spectral temporal maps in response to pilocarpine. (**D**) Quantification of total EEG power (seizure intensity) in response to pilocarpine. Open circles indicate each individual value. Horizontal bars indicate mean value (mean ± S.E.M.; *n* = 7, respectively). (**E**–**G**) Effects of CDDO-Me on neuronal death and 4-hydroxy-2-nonenal (4-HNE) signals following SE. CDDO-Me mitigates CA1 neuronal damage, but not hilus interneurons, although it reduces 4-HNE signals in both neurons. (**E**) Representative photos of double immunofluorescent staining for terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and 4-HNE. (**F**) Quantifications of the number of TUNEL positive neurons and (**G**) and the fluorescent intensity of 4-HNE in response to pilocarpine. Open circles indicate each individual value. Horizontal bars indicate mean value (mean ± S.E.M.; \* *p* < 0.05 vs. vehicle; *n* = 7, respectively).

#### *3.2. CDDO-Me Induces Mitochondrial Fission in CA1 Neuron and PV Cells without Altering LONP1 Expression*

As mentioned previously, the dysfunctions of mitochondrial dynamics induced by SE lead to apoptosis and programmed necrosis in PV cells and CA1 neurons, respectively [7,8,10,16–18]. Thus, we evaluated the effects of CDDO-Me on mitochondrial dynamics in control- and post-SE animals. In control animals, CDDO-Me reduced mitochondrial length in CA1 neurons without altering mitochondrial LONP1 expression (*p* < 0.05 vs. vehicle; Figure 2A–D). Following SE, CDDO-Me significantly inhibited mitochondrial elongation, but not mitochondrial LONP1 over-expression, in these neurons (*p* < 0.05 vs. control animals; Figure 2A–D). Similarly, CDDO-Me led to mitochondrial fragmentation in PV cells in control animals without changing mitochondrial LONP1 expression (*p* < 0.05 vs. vehicle; Figure 3A–D), while it could not affect SE-induced excessive mitochondrial fission and LONP1 over-expression (Figure 3A–D). These findings indicate that CDDO-Me may facilitate mitochondrial fission in CA1 neurons as well as PV cells, which reversely influences neuronal death in both neuronal subpopulations following SE [10,25].

**Figure 2.** Effects of CDDO-Me on mitochondrial length and LONP1 expression in CA1 neurons. CDDO-Me reduces mitochondrial length in CA1 neurons of (**A**) control- and (**B**) post-SE animals. CDDO-Me does not affect the increased LONP1 expression following SE. (**A**,**B**) Representative photos of LONP1, mitochondria (Mito) and nuclei (DAPI) in CA1 neurons. (**C**,**D**) Quantifications of (**C**) the mitochondrial length and (**D**) LONP1 fluorescent intensity in CA1 neurons. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate SEM (\*,# *p* < 0.05 vs. vehicle and control animals, respectively; *n* = 7, respectively).

**Figure 3.** Effects of CDDO-Me on mitochondrial length and LONP1 expression in parvalbumin (PV) cells. CDDO-Me reduces mitochondrial length in CA1 neurons of (**A**) control- and (**B**) post-SE animals. CDDO-Me does not affect the increased LONP1 expression following SE. (**A**,**B**) Representative photos of LONP1, mitochondria (Mito) and PV. (**C**,**D**) Quantifications of (**C**) the mitochondrial length and (**D**) LONP1 fluorescent intensity in PV cells. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate SEM (\*,# *p* < 0.05 vs. vehicle and control animals, respectively; *n* = 7, respectively).

#### *3.3. CDDO-Me Enhances DRP1-S616 Phosphorylation in Control- and Post-SE Animals*

Since CDDO-Me exerted mitochondrial fragmentation in control and post-SE animals, we explored whether CDDO-Me influences the machinery molecules of mitochondrial dynamics. SE did not affect OPA1 andMFN1/2 expression levels (Figure 4A–D). Consistent with our previous studies [8,10,16–18,25], DRP1 expression was reduced to 0.68-fold of the control level following SE (*p* < 0.05 vs. control animals; Figure 4A,E). Furthermore, SE decreased the DRP1-S616 phosphorylation level to 0.67-fold of the control level (*p* < 0.05 vs. control animals; Figure 4A,F), while it did not affect the DRP1-S637 phosphorylation level (Figure 4A,G). These findings indicate a reduction in the DRP1-S616/S637 phosphorylation ratio following SE. In control animals, CDDO-Me did not influence OPA1, MFN1/2 and DRP1 expression levels (Figure 4A–E). However, CDDO-Me increased the DRP1-S616, but not S637, phosphorylation level to 1.5-fold of the vehicle level (*p* < 0.05 vs. vehicle; Figure 4A,F,G). In addition, CDDO-Me attenuated the down-regulation of DRP1 expression and its S616 phosphorylation level induced by SE. (*p* < 0.05 vs. control animals and vehicle, respectively; Figure 4A,F,G). Therefore, CDDO-Me increased the DRP1-S616/S637 phosphorylation ratio under physiological- and post-SE conditions. Since S616 site phosphorylation of DRP1 facilitates mitochondrial fission [26], our findings sugges<sup>t</sup> that CDDO-Me may result in mitochondrial fragmentation via the enhancement of DRP1-S616 phosphorylation.

**Figure 4.** Effects of CDDO-Me on the expression and phosphorylation of mitochondrial dynamics-related molecules. CDDO-Me increases only the dynamin-related proteins 1 (DRP1)-S616 phosphorylation level under physiological- and post-SE conditions. (**A**) Representative images for western blots of optic atrophy 1 (OPA1), mitofusin 1 (MFN1), MFN2, DRP1, phospho (p)-DRP1-S616 and pDRP1-637 in the hippocampal tissues. (**B**–**G**) Quantifications of OPA1, MFN1 and MFN2, DRP1, pDRP1-S616 and pDRP1-637 levels. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate SEM (\*,# *p* < 0.05 vs. control animals and vehicle, respectively; *n* = 7, respectively).

#### *3.4. CDDO-Me Increases ERK1*/*2 and JNK Activities in the Hippocampus*

DRP1 phosphorylations are modulated by the ERK1/2, PKA and JNK signal pathways [26–30]. Furthermore, CDDO-Me influences ERK1/2 [20,31] and JNK phosphorylations [32,33]. Therefore, we evaluated the effects of CDDO-Me on the activities (phosphorylations) of protein kinases that are involved in DRP1 phosphorylations. In the present study, SE significantly reduced ERK1/2 and JNK phosphorylation levels without changing their expression (*p* < 0.05 vs. control animals, Figure 5A,B,C,F,G), while it did not affect the expression and phosphorylation levels of PKA catalytic and regulatory subunits (Figure 5A,D,E). In control animals, CDDO-Me did not change the protein expression levels of ERK1/2, PKA and JNK (Figure 5A,B,D,F). However, CDDO-Me increased the phosphorylation levels of ERK1/2 and JNK, but not PKA catalytic and regulatory subunits (*p* < 0.05 vs. vehicle; Figure 5A,C,G). Furthermore, CDDO-Me mitigated SE-induced reductions in ERK1/2 and JNK phosphorylations (*p* < 0.05 vs. vehicle; Figure 5A,C,G).

Protein phosphatases also regulate mitochondrial dynamics via DRP1 dephosphorylations. Indeed, protein phosphatase (PP) 2B (calcineurin) facilitates mitochondrial fission by dephosphorylating DRP1-S637 [34]. Thus, we validated whether CDDO-Me affects PP activities in the hippocampus. In control animals, CDDO-Me did not alter the protein expressions of PP1, PP2A and PP2B and their phosphorylation levels (Figure 6A–G). SE significantly reduced PP2B (not PP1 and PP2A) phosphorylation levels in the hippocampus, indicating an increase in its activity (*p* < 0.05 vs. control animals, Figure 6A,C,E,G), which was not affected by CDDO-Me (Figure 6A,C,E,G). Taken together, our findings indicate that CDDO-Me may regulate DRP1 S616 phosphorylation by enhancing ERK1/2 and JNK activities, independent of PP activities.

**Figure 5.** Effects of CDDO-Me on expressions and phosphorylations of extracellular-signal-regulated kinase 1/2 (ERK1/2), protein kinase A (PKA) and c-Jun N-terminal kinase (JNK). CDDO-Me increases ERK1/2 and JNK phosphorylation levels, but not PKA catalytic (PKA-Cat) and regulatory (PKA-Reg) subunits under physiological- and post-SE conditions. (**A**) Representative images for western blot of ERK1/2, phospho (p)-ERK1/2, PKA, pPKA, JNK and pJNK in the hippocampal tissues. (**B**–**G**) Quantifications of ERK1/2, pERK1/2, PKA, pPKA, JNK and pJNK levels. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate SEM (\*,# *p* < 0.05 vs. control animals and vehicle, respectively; *n* = 7, respectively).

**Figure 6.** Effects of CDDO-Me on the expression and phosphorylation of protein phosphatase (PP) 1, PP2A and PP2B. CDDO-Me does not affect the expression and phosphorylation levels of protein phosphatase 1 (PP1), PP2A and PP2B under physiological- and post-SE conditions. (**A**) Representative images for western blot of PP1, phospho (p)-PP1, PP2A, pPP2A, PP2B and pPP2B in the hippocampal tissues. (**B**–**G**) Quantifications of PP1, pPP1, PP2A, pPP2A, PP2B and pPP2B levels. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate S.E.M. (\* *p* < 0.05 vs. control animals; *n* = 7, respectively).

#### *3.5. E*ff*ects of CDDO-Me on SE-Induced Neuronal Death and Mitochondrial Dynamics are Independent of LONP1 Activity*

In the present study, we found that CDDO-Me facilitated mitochondrial fission by increasing ERK1/2- and JNK-mediated DRP1-S616 phosphorylation. However, it was unclear whether these effects are relevant to the reduced LONP1 activity or the additional CDDO-Me actions. To directly elucidate the role of LONP1 in mitochondrial dynamics, we applied LONP1 siRNA prior to SE induction. In control animals, LONP1 knockdown effectively reduced LONP1 expression in the hippocampus (*p* < 0.05 vs. control siRNA; Figure 7A,B). Although LONP1 knockdown did not affect seizure susceptibility in response to pilocarpine (Figure 7C,D), it attenuated up-regulation of LONP1 expression induced by SE (*p* < 0.05 vs. control siRNA; Figure 7A,B). However, LONP1 siRNA exacerbated SE-induced neuronal damage in the CA1 neurons, hilus neurons and dentate granule cells, unlike CDDO-Me (*p* < 0.05 vs. control siRNA; Figure 7E,F). Co-treatment of CDDO-Me with LONP1 siRNA ameliorated only CA1 neuronal death, but not hilus neurons and dentate granule cells, following SE (*p* < 0.05 vs. LONP1 siRNA; Figure 7E,F). Furthermore, LONP1 siRNA did not influence the expression and phosphorylation of DRP1, ERK1/2, JNK and PP2B under physiological- and post-SE conditions (Figure 8A–J). LONP1 knockdown could not inhibit SE-induced mitochondrial elongation in CA1 neurons (Figure 8K,L). However, co-treatment of CDDO-Me with LONP1 siRNA abrogated mitochondrial elongation induced by SE (*p* < 0.05 vs. LONP1 siRNA; Figure 8K,L). These findings indicate that CDDO-Me-mediated mitochondrial fission may be independent of LONP1 activity, and that up-regulation of LONP1 may be an adaptive response to protect neurons from SE.

**Figure 7.** Effects of LONP1 siRNA on LONP1 expression, seizure activity and neuronal damage in response to pilocarpine. (**A**,**B**) Effect of LONP1 knockdown on LONP1 expression in response to pilocarpine. LONP1 siRNA effectively reduced LONP1 expression under physiological- and post-SE conditions. (**A**) Representative western blots of LONP1 expression. (**B**) Quantification of LONP1 expression based on western blot data. Open circles indicate each individual value. Horizontal bars indicate mean value (mean ± S.E.M.; \* *p* < 0.05 vs. control animals, respectively; *n* = 7). (**C**,**D**) Effect of LONP1 knockdown on seizure activity in response to pilocarpine. LONP1 siRNA does not influence seizure activity induced by pilocarpine. (**C**) Representative EEG traces and frequency-power spectral temporal maps in response to pilocarpine. (**D**) Quantification of total EEG power (seizure intensity) in response to pilocarpine. Open circles indicate each individual value. Horizontal bars indicate mean value (mean ± S.E.M.; *n* = 7, respectively). (**E**,**F**) Effects of LONP1 siRNA and co-treatment of CDDO-Me on neuronal death induced by pilocarpine. LONP1 knockdown deteriorates degenerations of CA1 neurons, hilus interneurons (**H**) and dentate granule cells (DGC). Co-treatment of CDDO-Me attenuated CA1 neuronal death induced by SE. (**E**) Representative photos of FJB positive degenerating neurons. (**F**) Quantification of the number of FJB positive neurons in response to pilocarpine (mean ± S.E.M.; \*,# *p* < 0.05 vs. control siRNA and LONP1 siRNA, respectively; *n* = 7, respectively).

**Figure 8.** Effects of LONP1 knockdown on expression/phosphorylation of DRP1, ERK1/2, JNK and PP2B, and mitochondrial length in CA1 neurons. LONP1 siRNA does not affect the expression and phosphorylation levels of DRP1, ERK1/2, JNK and PP2B under physiological- and post-SE conditions. In addition, LONP1 knockdown does not influence SE-induced mitochondrial elongation in CA1 neurons, although it reduces LONP1 expression. However, co-treatment of CDDO-Me ameliorates mitochondrial elongation. (**A**) Representative images for western blots of DRP1, phospho (p)-DRP1-S616, pDRP1-637, ERK1/2, pERK1/2, JNK, pJNK, PP2B and pPP2B in the hippocampal tissues. (**B**–**J**) Quantifications of DRP1, pDRP1-S616, pDRP1-637, ERK1/2, pERK1/2, JNK, pJNK, PP2B and pPP2B levels. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate S.E.M. (\* *p* < 0.05 vs. control animals; *n* = 7, respectively). (**K**) Representative photos of mitochondria (Mito) and LONP1 in CA1 neurons following SE. (**L**) Quantification of the mitochondrial length in CA1 neurons. Open circles indicate each individual value. Horizontal bars indicate mean value. Error bars indicate S.E.M. (\*,# *p* < 0.05 vs. control siRNA and LONP1 siRNA, respectively; *n* = 7, respectively).
