A phosphodiesterase 4 (PDE4) inhibitor, amlexanox, reduces neuroinflammation and neuronal death after pilocarpine-induced seizure

Abstract

Epilepsy, a complex neurological disorder, is characterized by recurrent seizures caused by aberrant electrical activity in the brain. Central to this study is the role of lysosomal dysfunction in epilepsy, which can lead to the accumulation of toxic substrates and impaired autophagy in neurons. Our focus is on phosphodiesterase-4 (PDE4), an enzyme that plays a crucial role in regulating intracellular cyclic adenosine monophosphate (cAMP) levels by converting it into adenosine monophosphate (AMP). In pathological states, including epilepsy, increased PDE4 activity contributes to a decrease in cAMP levels, which may exacerbate neuroinflammatory responses. We hypothesized that amlexanox, an anti-inflammatory drug and non-selective PDE4 inhibitor, could offer neuroprotection by addressing lysosomal dysfunction and mitigating neuroinflammation, ultimately preventing neuronal death in epileptic conditions. Our research utilized a pilocarpine-induced epilepsy animal model to investigate amlexanox's potential benefits. Administered intraperitoneally at a dose of 100 ​mg/kg daily following the onset of a seizure, we monitored its effects on lysosomal function, inflammation, neuronal death, and cognitive performance in the brain. Tissue samples from various brain regions were collected at predetermined intervals for a comprehensive analysis. The study's results were significant. Amlexanox effectively improved lysosomal function, which we attribute to the modulation of zinc's influx into the lysosomes, subsequently enhancing autophagic processes and decreasing the release of inflammatory factors. Notably, this led to the attenuation of neuronal death in the hippocampal region. Additionally, cognitive function, assessed through the modified neurological severity score (mNSS) and the Barnes maze test, showed substantial improvements after treatment with amlexanox. These promising outcomes indicate that amlexanox has potential as a therapeutic agent in the treatment of epilepsy and related brain disorders. Its ability to combat lysosomal dysfunction and neuroinflammation positions it as a potential neuroprotective intervention. While these findings are encouraging, further research and clinical trials are essential to fully explore and validate the therapeutic efficacy of amlexanox in epilepsy management.

Keywords: Autophagy; Epilepsy; Lysosome; Neuro-inflammation; Phosphodiesterase4; cAMP.

Fig. 1

Study of Amlexanox's Impact on Seizure Frequency, Body Weight, and Mortality Rate in Pilocarpine-Induced Seizures. A. Experimental Timeline: Outlines the procedure involving pilocarpine administration and the subsequent monitoring of the seizure intensity at 5-min intervals until Stage 4 (rearing with forelimb clonus). Additionally, details the schedule for body weight measurements in the various groups (sham vehicle, sham amlexanox, seizure vehicle, and seizure amlexanox) following the seizure episodes. B. Seizure Assessment: Illustrates the frequency of the seizure evaluation, conducted every 5 ​min following pilocarpine administration, until Stage 4 was reached. Presents the duration taken by each group to reach Stage 4, highlighting that the average onset time for seizures in both the seizure vehicle and seizure amlexanox groups was around 30 ​min. C. Body Weight Comparison: Shows body weight data for each group (sham vehicle, sham amlexanox, seizure vehicle, seizure amlexanox) gathered after the occurrence of seizures. D. Mortality Rate Pie Chart: Displays the mortality rates in rats subjected to pilocarpine-induced seizures, with the blue segment representing survival and the purple segment indicating mortality. E. Survival Graph: Plots the weekly survival counts of the experimental animals in the model, categorized by week. F. Seizure Frequency Graph: Depicts the frequency of seizures observed in the model during the mNSS test, spanning from 4 ​h after induction to 14 days. The graph presents the mean frequency of seizures, noting a statistically significant difference (p ​< ​0.05) between the vehicle group and the amlexanox-treated group (ANOVA after repeated-measures testing, seizure frequency: time: F ​= ​2.877, p ​= ​0.001; group: F ​= ​30.202, p ​< ​0.001; interaction: F ​= ​2.179, p ​= ​0.009).

Fig. 2

Amlexanox's Effect on PDE4B Expression, Pro-Inflammatory Factors, and Zinc Levels in Lysosomes Following Seizures. A. PDE4B Staining: Illustrates phosphodiesterase-4B (PDE4B) expression (in red) 12 ​h after a seizure and in the sham groups. Neurons were visualized using DAPI, which was merged with PDE4B staining, highlighting the neuronal population. B. PDE4B Quantification Bar Graph: Compares the PDE4B levels between the vehicle and amlexanox groups following seizures. The analysis involved the sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle (n ​= ​5), and seizure amlexanox (n ​= ​5) groups. The graph presents the mean ​± ​S.E.M. data, showing a significant difference (p ​< ​0.05) between the vehicle and amlexanox groups, as determined by the Bonferroni post hoc test following the Kruskal–Wallis test (chi-square ​= ​13.834, df ​= ​3, p ​= ​0.003). C. Zinc Staining with FluoZin-3: Demonstrates staining with FluoZin-3 in the cornu ammonis 1 (CA1) area, performed 72 ​h after a seizure. LysoTracker was used in conjunction with FluoZin-3 staining. D. Zinc Quantification Bar Graph: Displays the comparison of the zinc levels in the vehicle and amlexanox groups following seizures, involving the seizure vehicle (n ​= ​4) and seizure amlexanox (n ​= ​4) groups. The graph shows the mean ​± ​S.E.M. data, with a significant difference (p ​< ​0.05) between the groups as per the Mann–Whitney U test (z ​= ​2.309, p ​= ​0.029). E-G. Inflammatory Factor Staining: Displays the interleukin-6 (IL-6, in green) and tumor necrosis factor-α (TNF-α, in red) staining conducted 12 ​h following seizures and in sham groups. DAPI was merged with IL-6 and TNF-α staining. Inflammatory Factor Quantification Bar Graph: Compares the levels of IL-6 and TNF-α between the vehicle and amlexanox groups following seizures. The analysis included the sham vehicle (n ​= ​3), sham amlexanox (n ​= ​3), seizure vehicle (n ​= ​5), and seizure amlexanox (n ​= ​5) groups. The data are the mean ​± ​S.E.M., highlighting a significant difference (p ​< ​0.05) between the groups as established by the Bonferroni post hoc test after the Kruskal–Wallis test (IL-6: chi-square ​= ​13.412, df ​= ​3, p ​= ​0.004; TNF-α: chi-square ​= ​13.412, df ​= ​3, p ​= ​0.004). Scale bar corresponds to 20 ​μm in all images in Fig. 2.

Fig. 3

Amlexanox Enhances Autophagy and Lysosomal Function While Reducing Neuroinflammation. A-C. LAMP2 and LC3B Expression Analysis via Western Blot: This segment illustrates the assessment of the lysosome-associated membrane protein 2 (LAMP2) and light chain 3B (LC3B) levels, indicators of lysosomal function and autophagy. The displayed results show the expression levels of LAMP2 and LC3B in various experimental samples. The accompanying bar graphs quantify these results, indicating the relative protein expression in comparison to the control group. A significant difference in protein levels was observed for both LAMP2 (Bonferroni post hoc test after Kruskal–Wallis test, chi-square ​= ​10.063, df ​= ​3, p ​= ​0.018) and LC3B (Bonferroni post hoc test after Kruskal–Wallis test, chi-square ​= ​10.406, df ​= ​3, p ​= ​0.015). D-E. COX-2 Expression Analysis via Western Blot: These sections cover the assessment of cyclooxygenase-2 (COX-2) expression, a marker for neuroinflammation. The results present the COX-2 expression levels in the experimental samples. The corresponding bar graph quantifies these results, showing the relative expression of COX-2 compared to the control group. The statistical analysis revealed a significant difference in COX-2 levels (Bonferroni post hoc test after Kruskal–Wallis test, chi-square ​= ​11.709, df ​= ​3, p ​= ​0.008).

Fig. 4

Amlexanox Decreases Neuronal Death after Pilocarpine-Induced Seizures. A. NeuN Staining for Neuronal Survival: Depicts the use of neuronal nuclear protein (NeuN) staining to evaluate neuronal survival in the subiculum, CA1, cornu ammonis 3 (CA3), and hilus regions one week after pilocarpine-induced seizures, as well as in the sham group. The fluorescent images display the levels of NeuN-positive neurons in these specific brain regions. A scale bar indicating 100 ​μm is included, providing a reference for the size of the neuronal structures captured in the images. B-E. Quantitative Analysis of NeuN-Positive Neurons: These bar graphs compare the numbers of NeuN-positive neurons in different groups: sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle group (n ​= ​9), and seizure amlexanox (n ​= ​9). Measurements were taken one week after seizures and in the sham group. Data are presented as the mean ​± ​standard error of the mean (S.E.M.). The statistical analysis highlighted significant differences (p ​< ​0.05) between the groups as follows: subiculum (Bonferroni post hoc test after Kruskal-Wallis test, chi-square ​= ​18.067, df ​= ​3, p ​< ​0.001); CA1 (chi-square ​= ​22.138, df ​= ​3, p ​< ​0.001); CA3 (chi-square ​= ​21.476, df ​= ​3, p ​< ​0.001); hilus (chi-square ​= ​15.005, df ​= ​3, p ​= ​0.002).

Fig. 5

Amlexanox Alleviates Glial Cell Activation Following Seizures. Panels A–C show staining for glial fibrillary acidic protein (GFAP, green) to highlight astrocytes in the CA1 region post-seizure and in the sham group. Complement 3 (C3, red) staining was used to identify activated astrocytes in both groups. The fluorescent images display the localization of GFAP, C3, and 4′,6-diamidino-2-phenylindole (DAPI, marking cell nuclei), accompanied by a scale bar of 100 ​μm. The accompanying bar graphs (B and C) illustrate the intensity of GFAP and C3 staining in the CA1 region. The statistical analysis revealed significant differences between the groups (p ​< ​0.05), as shown by the Bonferroni post hoc test following a Kruskal–Wallis test (GFAP—chi-square ​= ​24.108, degrees of freedom (df) ​= ​3, p ​< ​0.001; C3—chi-square ​= ​21.375, df ​= ​3, p ​< ​0.001). Panels D–F feature staining for ionized calcium-binding adapter molecule 1 (Iba-1, green) to visualize microglia in the CA1 region post-seizure and in the sham group. Staining for cluster of differentiation 68 (CD68, red) was used to detect activated microglia. The images display Iba-1, CD68, and DAPI localization, with a scale bar of 100 ​μm. The bar graphs (E and F) present the intensity of the Iba-1 and CD68 staining in the CA1 area. The statistical analysis indicated significant differences between groups (p ​< ​0.05), as confirmed by the Bonferroni post hoc test following the Kruskal–Wallis test (Iba-1—chi-square ​= ​19.389, df ​= ​3, p ​< ​0.001; CD68—chi-square ​= ​22.938, df ​= ​3, p ​< ​0.001).

Fig. 6

Amlexanox Alleviates Oxidative Stress and Preserves Microtubule Integrity Following Seizures. A-E. Visualization of Microtubules with MAP2 Staining: Showcases microtubule-associated protein 2 (MAP2) staining (green fluorescence) across the hippocampus region, including the subiculum, CA1, CA3, and dentate gyrus (DG). The images display the distribution of MAP2 (green) alongside DAPI (blue), which indicates cell nuclei. A 100 ​μm scale bar provides a reference for the size of the structures. The bar graph depicts the percentage area of MAP2 staining in the entire hippocampus, including the sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle (n ​= ​9), and seizure amlexanox groups (n ​= ​9). The statistical analysis, as shown by the Bonferroni post hoc test after the Kruskal–Wallis test, indicated significant differences between the groups in CA1 (chi-square ​= ​22.32, df ​= ​3, p ​< ​0.001); data for other regions (subiculum—chi-square ​= ​23.916, df ​= ​3, p ​< ​0.001, CA3—chi-square ​= ​21.283, df ​= ​3, p ​< ​0.001, DG—chi-square ​= ​20.817, df ​= ​3, p ​< ​0.001) are indicated as significant where relevant. F-J. ROS Detection Using 4-HNE Staining: Shows reactive oxygen species (ROS) levels assessed through 4-hydroxyl-2-nonenal (4-HNE) staining in the entire hippocampus, including the subiculum, CA1, CA3, and DG. The corresponding bar graph illustrates the intensity of 4-HNE staining across the hippocampus, including the sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle (n ​= ​9), and seizure amlexanox groups (n ​= ​9). Mean values ​± ​S.E.M. are shown. Statistical significance (p ​< ​0.05) is highlighted between the groups (Bonferroni post hoc test after Kruskal–Wallis test: subiculum—chi-square ​= ​19.342, df ​= ​3, p ​< ​0.001; CA1—chi-square ​= ​20.563, df ​= ​3, p ​< ​0.001; CA3—chi-square ​= ​19.078, df ​= ​3, p ​< ​0.001; DG—chi-square ​= ​18.816, df ​= ​3, p ​< ​0.001).

Fig. 7

Amlexanox Mitigates Blood–Brain Barrier Disruption Following Seizures. A. Hippocampal Immunostaining: Features micrographs of the hippocampus stained with anti-rat IgG to indicate blood–brain barrier (BBB) disruption. The scale bar represents a length of 100 ​μm, providing a reference for the size of the structures observed. B. IgG Leakage Quantification: The bar graph quantifies the extent of IgG leakage in the hippocampus, an indicator of BBB integrity. The dataset includes the sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle (n ​= ​9), and seizure amlexanox groups (n ​= ​9), with values shown as the mean ​± ​S.E.M. The statistical analysis confirmed the significant difference (p ​< ​0.05) between the groups (Bonferroni post hoc test after Kruskal–Wallis test, chi-square ​= ​22.029, df ​= ​3, p ​< ​0.001), demonstrating the effectiveness of amlexanox in reducing BBB disruption following seizures.

Fig. 8

Amlexanox's Effects on Cognitive and Neuological Function Following Seizures. A. Experimental Procedure: The study began with the modified neurological severity score (mNSS) test 4 ​h after seizure induction, continuing through day 14. From days 14–28, the Barnes maze test was conducted. Amlexanox was administered daily at a dose of 100 ​mg/kg via intraperitoneal injection until the end of the experiment. B. mNSS Scores: These were recorded from 4 ​h post-seizure to day 14, on a scale ranging from 0 to 18 points. Statistical significance was determined by ANOVA with repeated measures (mNSS: time: F ​= ​37.795, p ​< ​0.001; group: F ​= ​11.04, p ​= ​0.004; interaction: F ​= ​2.316, p ​= ​0.005). ∗p ​< ​0.05. C. Barnes Maze Configuration: The maze consisted of a disk with a diameter of 122 ​cm, featuring an escape hole that was 7 ​cm in diameter. D. Barnes Maze Execution: Conducted from days 14–28 post-seizure with a time limit of 2 ​min (120s) per trial; failures were recorded as 120s. The statistical analysis involved ANOVA with repeated measures (Barnes maze: time: F ​= ​12.627, p ​< ​0.001; group: F ​= ​1.634, p ​= ​0.016; interaction: F ​= ​16.147, p ​< ​0.001). ∗p ​< ​0.05. E. Neuronal Survival via NeuN Staining: Live neurons were stained with NeuN in the subiculum, CA1, CA3, and hilus areas one week after pilocarpine-induced seizures and in the sham group. Fluorescent images show NeuN-positive neurons with a 100 ​μm scale bar. F–I. Quantification of NeuN-Positive Neurons: The bar graphs compare the NeuN-positive neuron counts in the vehicle and amlexanox treatment groups post-seizure and in the sham group. The sample sizes include the sham vehicle (n ​= ​5), sham amlexanox (n ​= ​5), seizure vehicle (n ​= ​2), and seizure amlexanox (n ​= ​7) groups. Values are expressed as the mean ​± ​S.E.M. Significance was assessed using the Bonferroni post hoc test following the Kruskal–Wallis test (subiculum: chi-square ​= ​10.77, df ​= ​3, p ​= ​0.013; CA1: chi-square ​= ​14.574, df ​= ​3, p ​= ​0.002; CA3: chi-square ​= ​14.498, df ​= ​3, p ​= ​0.002; hilus: chi-square ​= ​10.509, df ​= ​3, p ​= ​0.015). ∗p ​< ​0.05.

Fig. 9

Hypothetical Framework of the Current Study. A. Pilocarpine-Induced Seizure Schematic: This diagram provides a detailed representation of the process and critical stages in pilocarpine-induced seizures. It outlines the key steps and elements involved in the seizure induction mechanism. B. Amlexanox Administration Mechanism: This illustrates the role of amlexanox, a PDE4B inhibitor, highlighting its potential anti-inflammatory and neuroprotective effects following seizure occurrence. The diagram explains the conceptual mechanism of action of amlexanox in the context of seizure treatment and brain protection.

Additional File 1Original Western Blot Images. A. Protein Marker: Showcases the use of T&I™ ACCU Prestained Protein Marker for Western blot reference. B. LAMP2 Western Blot: Pzx€zresents the original Western blot image of lysosome-associated membrane protein 2 (LAMP2). Image no. 2 from this series was selected for representation in Figure 3. C. LC3B Western Blot: Displays the original Western blot image of light chain 3B (LC3B). Image no. 1 from this collection is used in Figure 3. D. β-Actin Western Blot: Features the original Western blot image of β-actin, with image no. 2 chosen for depiction in Figure 3. E, F. COX-2 and β-Actin Western Blots: Show the original Western blots for cyclooxygenase-2 (COX-2) and β-actin. Image no. 1 from this set is utilized in Figure 3.