Neuroprotective potential of isofraxidin: Alleviating parkinsonian symptoms,inflammation and microglial activation

Abstract

Background

Parkinson’s disease (PD) is one of the most common neurodegenerative disorders. Previous research has confirmed that isofraxidin can reduce macrophage expression and inhibit peripheral inflammation. However, its effects on the central nervous system remain underexplored.

Objective

This study aims to determine whether isofraxidin offers protective effects against PD.

Methods

To assess the effects of isofraxidin, motor performance changes in LPS-induced PD mice were evaluated using rotarod, pole-climbing, and beam-walking tests. Striatal damage was examined through [¹⁸F]fluorodeoxyglucose ([¹⁸F]FDG) positron emission tomography (PET) imaging, and dopaminergic neurotoxicity was assessed using tyrosine hydroxylase (TH) staining. Microglial accumulation and activation were monitored with Iba-1 staining, while LPS-induced inflammation was examined via TNF-α and IL-1β staining.

Results

Isofraxidin pre-treatment significantly improved LPS-induced motor dysfunction, as evidenced by better performance in the rotarod, pole-climbing, and beam-walking tests. [¹⁸F]FDG PET imaging showed that isofraxidin restored glucose uptake in the striatum, countering LPS-induced damage. Furthermore, Iba-1 staining revealed that isofraxidin markedly inhibited LPS-induced microglial activation and accumulation. TNF-α and IL-1β staining indicated a reduction in inflammation with isofraxidin treatment. Additionally, TH staining supported the neuroprotective role of isofraxidin on dopaminergic neurons.

Conclusions

Isofraxidin exhibits notable neuroprotective properties by mitigating LPS-induced parkinsonian behaviors, microglial activation, inflammation, and dopaminergic neuron damage. These results highlight isofraxidin’s potential as a therapeutic intervention for PD.

Plain Language Summary

What’s the study about? Parkinson's disease is a common brain disorder, and we wanted to see if a substance called isofraxidin could help. Isofraxidin is known to reduce inflammation in the body, but its effects on the brain haven't been well studied. What did we do? We tested isofraxidin on mice with Parkinson's disease-like symptoms caused by a substance called LPS. We checked the mice's movement abilities using various tests, looked at their brain using special imaging techniques, and examined specific brain cells and markers related to Parkinson's disease. What did we find? Isofraxidin improved the mice's movement abilities, reduced brain damage, and lowered inflammation in the brain. It also protected specific brain cells affected by Parkinson's disease. What does it mean? These results suggest that isofraxidin could be a helpful treatment for Parkinson's disease by reducing symptoms and protecting the brain. Further research could explore its potential for treating this condition in humans.

Introduction

Parkinson’s disease (PD) stands as one of the most prevalent neurodegenerative disorders, second only to Alzheimer’s disease. An epidemiological study indicated that the prevalence of PD increases with ages, particularly affecting the population after the age of 60. ¹ A research study estimated that in 2017, around one million people in the U.S. were diagnosed with PD, resulting in a total economic burden of $51.9 billion. ²

PD is characterized by the selective loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNpc), leading to movement disorders such as resting tremors, rigidity, akinesia or bradykinesia, and postural instability. ³ The pathogenic mechanism of PD remains unclear and is generally considered to result from the interaction between age, genetics, and the environment. ⁴ In addition, inflammation has been implicated in the pathogenesis and progression of PD, ⁵ with microglial activation playing a significant role. ⁶

LPS is a significant component of the outer membrane in Gram-negative bacteria. Over the past 20 years, research has demonstrated that LPS-induced inflammation, through the binding of LPS to microglial TLR4 receptor, can elicit pathological PD characteristics in animals. This includes increased activation of microglia and the loss of DA neurons in the nigrostriatal system. ⁷ Microglia, classified as parenchymal macrophages in the CNS, are implicated in neuroinflammation through the secretion of cytokines such as TNF-α, IL-6, and IL-1. ⁸ Growing evidence suggests that activated microglia may play a significant role in neurodegenerative processes, potentially acting as primary instigators of conditions like PD. ⁶ Current medications for PD primarily focus on addressing the imbalance between the endogenous dopamine and acetylcholine. However, these medications only alleviate symptoms without effectively addressing microglial inflammatory responses or reducing damage to DA neurons. Therefore, more research is needed to identify fundamental treatments or preventive measures for PD progression.

Isofraxidin, a coumarin compound found in medicinal plants such as Sarcandra glabra and Acanthopanax senticosus, has well-documented antibacterial, antioxidant, and anti-inflammatory properties. Research has shown that isofraxidin exerts anti-inflammatory effects in both in vitro and in vivo models. In particular, studies using an LPS-induced mouse peritoneal macrophage model demonstrated that isofraxidin significantly reduced the levels of the inflammatory cytokine TNF-α. This reduction is linked to the inhibition of the MAPK signaling pathway, as indicated by the decreased phosphorylation of p38 and ERK1/2 proteins. ⁹ Isofraxidin has also demonstrated the ability to target the TLR4/MD-2 axis in macrophages involved in osteoarthritis. ¹⁰ Other studies have indicated that pre-treatment with isofraxidin prior to LPS challenge could reduce mortality, body weight loss, organ coefficient, and histopathological changes. It also lowered serum levels of NF-κB, NO, and IL-6, along with decreasing TNF-α production in the liver. ¹¹ Isofraxidin’s anti-inflammatory effects were further demonstrated in an LPS-induced acute lung injury study, where it reduced TNF-α levels and macrophage expression in lung tissues. ¹² Additionally, isofraxidin exhibited protective effects against IL-1β-induced inflammation in intervertebral disc degeneration. ¹³ Although these studies all emphasized the remarkable anti-inflammatory effects of isofraxidin and its ability to modulate peripheral macrophages and regulate pro-inflammatory cytokines, it remains to be determined whether isofraxidin has similar inhibitory effects on microglia in the CNS.

In light of these findings, this study hypothesizes that isofraxidin could significantly mitigate the inflammatory response of microglia, thereby reducing DA neuronal damage and improving dyskinesia in PD. To confirm this hypothesis, we conducted rotarod, pole-climbing, and beam-walking tests to evaluate the behavioral performance of mice. [¹⁸F]FDG-PET imaging and immunohistochemistry (IHC) staining of TH in the striatum were employed to explore the neuroprotective effects of isofraxidin. IHC staining of Iba-1, TNF-α and IL-1β was performed to assess the extent of microglial activation and inflammation within the striatum.

Materials and Methods

Materials

Isofraxidin was purchased from MedChemExpress (catalogue number: HY-N0774), LPS (Salmonella enterica, catalogue number: SI-L9764), Novolink Polymer Detection System (catalogue number: RE7140-K), and chloral hydrate (catalogue number: 15307-500G-R) were purchased from Sigma-Aldrich (St. Louis, MO, USA). The anti-Iba-1 antibody (catalogue number: GTX632426) was purchased from Abcam (Cambridge, UK), and anti-tyrosine hydroxylase antibody (TH, catalogue number: AB152) was obtained from Millipore. The anti-TNF-α antibody (catalogue number: 60291-1-lg) was purchased from Proteintech, and the anti-IL-1β antibody (catalogue number: 12242) was purchased from Cell Signaling Technology. Isoflurane (Attane^®) was procured from Panion & Biotech (Taoyuan, Taiwan), and sodium chloride injection (catalogue number: XJ0606) was obtained from TAI YU (Hsinchu, Taiwan). Pure water for the experiment was produced by the Milli-Q Water Purification System (Millipore, Bedford, MA, USA). [¹⁸F]FDG was provided by Tri-Service General Hospital PET center and purchased from Global Medical Solution Taiwan, Ltd.

Animals

The animals utilized in this study were 11-week-old male C57BL/6 mice, weighing approximately 22 to 25 g, and were procured from BioLASCO Taiwan Co., Ltd., Taipei, Taiwan. These mice were housed at the National Defense Medical Center (NDMC), with three to four mice per cage, in a 12-hour light/dark cycle (daylight beginning at 7:00 AM). The room had a controlled temperature (23 ± 2°C) and humidity, and the mice were given unrestricted access to food and water. All experimental procedures were conducted from August 2022 to February 2024 in accordance with the Institutional Animal Care & Use Committee (IACUC-23-060) of NDMC.

Experiment 1: Dosage Screening of Isofraxidin for Anti-Parkinsonism Effects

In Experiment 1 (Figure 1(A)), mice were randomly assigned to five groups to evaluate the anti-parkinsonism effects of different isofraxidin dosage regimens. Rotarod training and baseline measurements were conducted from day 1 to day 4, with testing performed 7 days after stereotaxic surgery:

(1) Sham group (n = 3): Mice received an intra-striatal injection of normal saline (4 μL) on day 10 and oral administration of a carboxymethylcellulose (CMC) solution (0.5%, w/v) from day 8 to day 17 (0.1 mL/10 g body weight).

(2) LPS group (n = 3): Mice were injected with LPS (4 μL, 5 μg/μL) into the striatum on day 10 and orally received CMC solution (0.5%, w/v) following the same schedule.

(3) Isofraxidin (5 mg/kg, n = 3): Mice received LPS injection as above, followed by oral isofraxidin-CMC solution (0.5 mg/mL) from day 8 to day 17 (0.1 mL/10 g body weight).

(4) Isofraxidin (10 mg/kg, n = 3): Mice received the LPS injection, then isofraxidin-CMC solution (1 mg/mL) orally, administered in the same manner as above.

(5) Isofraxidin (20 mg/kg, n = 3): Mice received the LPS injection, followed by oral isofraxidin-CMC solution (2 mg/mL) according to the same schedule.

Figure 1.

Experimental design and results of isofraxidin dosage screening and its impact on microglial inflammatory response in an LPS-induced Parkinsonism model. (A) Experiment 1: dosage screening of isofraxidin for anti-parkinsonism effects. Mice were divided into different groups to receive varying doses of isofraxidin (5, 10, or 20 mg/kg) or vehicle control (CMC) following an LPS injection (5 μg/μL, total 4 μL) into the striatum on day 10. Isofraxidin or CMC was administered orally from day 8 to day 17. Rotarod training was conducted from day 1 to day 3, followed by baseline assessment on day 4. The rotarod test was then repeated on day 17 to evaluate motor performance post-treatment. (B) Experiment 2: assessment of isofraxidin’s impact on microglial inflammatory response in an LPS-induced Parkinsonism model. The timeline illustrates the sequence of behavioral testing (rotarod, pole-climbing, and beam-walking), followed by [¹⁸F]FDG PET imaging. Mice received stereotaxic injections (NS or LPS) into the striatum on day 10, followed by oral administration of isofraxidin (20 mg/kg) or CMC solution for 20 days (from day 8 to day 27). Behavioral tests were conducted at multiple time points, and [¹⁸F]FDG PET imaging was performed on day 43. The mice were sacrificed on day 44 for IHC staining to assess microglial activation. (C) Results of experiment 1: Rotarod performance post-isofraxidin treatment. The graph shows rotarod performance time (seconds) for each group at baseline and week 1 post-LPS injection. LPS significantly reduced performance compared to the sham group (^***P < 0.001). Isofraxidin treatment (10 and 20 mg/kg) significantly improved performance compared to the LPS group (^#P < 0.05, ^##P < 0.01, ^###P < 0.001), with the 20 mg/kg dose showing the most pronounced effect. CMC: carboxymethyl cellulose; NS: normal saline; LPS: lipopolysaccharide; PET: positron emission tomography; IHC: immunohistochemistry; pink arrow: pole-climbing baseline and four tests; red arrow: rotarod baseline and four tests.

Experiment 2: Assessment of Isofraxidin’s Impact on Microglial Inflammatory Response in an LPS-Induced Parkinsonism Model

Mice were randomly divided into four groups, and the complete experimental procedure is outlined in Figure 1(B):

(1) Sham group (n = 13): Received a 4 μL injection of normal saline (NS) into the striatum on day 10, along with oral administration of CMC solution (0.5%, w/v) from day 8 to day 27.

(2) LPS group (n = 15): Received a 4 μL injection of LPS into the striatum on day 10, with oral CMC administration from day 8 to day 27.

(3) ISO + LPS group (n = 12): Received a 4 μL injection of LPS into the striatum on day 10, along with oral administration of isofraxidin (20 mg/kg) from day 8 to day 27.

(4) ISO group (n = 14): Received a 4 μL injection of NS into the striatum on day 10, with oral administration of isofraxidin (20 mg/kg) from day 8 to day 27.

LPS – Striatum – Stereotaxic Surgery

Before surgery, mice were deeply anesthetized and had their hair shaved. They were then positioned on a stereotaxic apparatus, and the scalp was incised. Using the bregma as the reference point, coordinates for the striatum location were determined. The first and second coordinates were set at anterior/posterior +1.18 mm, medial/lateral ±1.5 mm, and dorsal/ventral -3.5 mm. The third and fourth coordinates were set at anterior/posterior -0.34 mm, medial/lateral ±2.5 mm, and dorsal/ventral -3.2 mm. Holes were then drilled accordingly using an electric drill. ¹⁴ LPS solution (5 μg LPS dissolved in 1 μL NS) was injected into each hole at a controlled rate of 0.5 μL per minute. After injection, a 5-minute waiting period was observed before slowly retracting the syringe to minimize material displacement. The drilled holes were sealed using tooth powder. The incised scalp was affixed using tissue adhesive and treated with povidone-iodine, antibiotic, and antihistamine ointments to complete the surgical procedure. In the sham and ISO groups, normal saline (NS) was substituted for the LPS solution (please refer to the Supplemental Materials and Methods for detailed procedures).

Rotarod Tests

The rotarod apparatus (UGO BASILE, Model 7700) was used to evaluate balance and motor coordination in mice. Mice underwent the following procedure during the training phase. Initially, they were placed on the stationary roller and allowed to stand calmly. Subsequently, the roller’s speed was gradually increased in increments of 5 rpm, 20 rpm, 24 rpm, 28 rpm, 32 rpm and 40 rpm for 1 minute respectively. Each mouse underwent this training regimen three times, with a minimum rest period of 30 minutes between consecutive trials.

During testing phase, the mice were placed on the stationary roller and allowed to stand calmly. The roller then accelerated from 5 rpm to 50 rpm over 300 seconds, with a maximum recording time of 300 seconds or until the mice fell off. Testing sessions were conducted during nocturnal hours, with each mouse undergoing three trials, separated by at least 30 minutes of rest. The average of the three trials was calculated for analysis, with slight modifications from prior studies. ¹⁴

Beam-Walking Tests

The beam-walking test assessed the dynamic balance of mice using a rectangular balance beam, 100 cm in length and 12 mm in width. A home cage was placed at one end of the beam as the destination, with a soft cushion placed underneath to prevent falls. The mice were acclimated to the environment for one hour before testing began. The testing procedure involved placing the mouse on the beam with an initial starting point of 10 cm. A light source was used to stimulate the mice to move forward along the beam. The time taken by the mouse to traverse the length of the beam was recorded, enabling an assessment of the mouse’s balance and coordination abilities. A 2-day training phase was conducted to observe the mice's learning, with the mice gently prompted to move forward if needed. On the third day, baseline data were collected during the testing phase.

Following a 4-week period, 3-day training and testing sessions were conducted to observe potential behavioral changes post-surgery. Testing occurred in a dark room at night to maximize mouse activity. The testing duration was 60 seconds, with recordings capped at 60 seconds if traversal wasn’t completed. Each mouse underwent three trials with a minimum 10 minutes of rest between. The process was videotaped, and timing data were manually recorded for analysis, with slight adaptations from prior studies to suit experimental needs. ¹⁵

Pole-Climbing Tests

Bradykinesia levels in mice were evaluated using pole-climbing tests. A 50 cm long and 1 cm diameter wooden pole was vertically secured to the floor, its surface covered with two layers of gauze for stability and to prevent skidding. ¹⁶ In the pole-climbing tests, mice were placed at the top of the wooden pole and allowed to descend freely. The test concluded once the mouse’s forelimbs touched the lower end of the pole and the time taken was recorded. Each mouse underwent three trials, and then the average time was calculated for analysis.

[¹⁸F]FDG and Positron Emitting Tomography (PET)

[¹⁸F]FDG-PET imaging is a valuable technique for assessing neuronal dysfunction in neurodegenerative diseases by visualizing cerebral glucose metabolism, which reflects synaptic activity.^17,18 LPS injection into the striatum induces inflammation and cell death, altering glucose absorption and metabolism, and allowing specific patterns of deficit in the brains of PD mice to be identified. ¹⁹

Prior to imaging, mice fasted for 24 hours, and then they received intraperitoneal [¹⁸F]FDG injections. After 2 hours of free movement, mice were anesthetized with isoflurane/oxygen for the PET imaging using the BIOPET 105 imager. A 15-minute PET scan was conducted, and images were reconstructed using 3D-OSEM and analyzed with AMIDE software. Standardized uptake values (SUV) of the striatum were obtained after correcting for injected activity and mouse weight. Brain images were manually matched to a template MRI image for regional radioactivity measurement (please refer to the Supplemental Materials and Methods for detailed procedures).

IHC Staining

Brain samples were sliced into 10 and 20 μm sections and placed on glass slides. After an overnight wash with 0.1 M PBS on day 1, peroxidase and protein blocks were applied on day 2, followed by overnight incubation with primary antibodies. The primary antibodies used were ionized calcium binding adaptor molecule 1 (Iba-1, 1:400), tumor necrosis factor-α (TNF-α, 1:500), interleukin-1β (IL-1β, 1:200), and tyrosine hydroxylase (TH, 1:1000). On day 3, samples underwent four PBS washes and a 30-minute incubation with secondary antibody. After four additional PBS washes, a 5% DAB solution was applied for 1-2 minutes to induce brown coloration, followed by the application of mounting medium. Samples were observed and imaged using a Canon compound microscope system. Iba-1 quantification was done with Image Pro Plus software, while TNF-α, IL-1β, and TH were quantified using ImageJ (please refer to the Supplemental Materials and Methods for detailed procedures).

Statistical Analysis

The experimental results were presented as mean ± standard error of the mean (SEM). The Shapiro-Wilk test was utilized to assess whether the data adhered to a normal distribution. If the data demonstrated normal distribution, one-way and two-way analysis of variance (ANOVA) were employed. Subsequent post-hoc analysis was carried out using the Bonferroni method in cases of homogenous variance, while the Dunnett T3 method was used for cases of heterogeneous variance. When the data did not adhere to normal distribution, the Kruskal-Wallis test was employed. All statistical analyses were two-tailed tests, with a significance level (α) of 0.05 for the type I error probability.

Results

Dose-Dependent Effects of Isofraxidin on Rotarod Performance Following LPS Injection

In Experiment 1, a dose-screening test using one-way ANOVA with Bonferroni post hoc comparisons revealed that LPS injection significantly reduced rotarod performance in mice compared to the sham group (P < 0.001). Mice pretreated with 5 mg/kg of isofraxidin did not show significant improvement in motor function compared to the LPS group. However, pretreatment with 10 mg/kg and 20 mg/kg of isofraxidin significantly enhanced rotarod performance (P < 0.05 and P < 0.001, respectively) compared to the LPS group, with the 20 mg/kg dose showing the most pronounced protective effect (Figure 1(C)). Based on these findings, the 20 mg/kg dose of isofraxidin was selected for subsequent Experiment 2.

Isofraxidin Administration Prior to LPS Injection Mitigates Motor Dysfunction in Parkinsonism-Like Models

In Experiment 2, rotarod tests evaluated motor function and balance coordination by measuring the time mice remained on the spinning rod. The LPS group consistently showed lower fall times (approximately 150 seconds) compared to the sham group and ISO groups (approximately 300 seconds) from week 1 to week 4 post-LPS injection. Importantly, the ISO + LPS group exhibited significantly improved fall latency (average 250 seconds) compared to the LPS group. Motor performance changes are illustrated in Figure 2(A), a two-way repeated measures ANOVA was conducted to assess the impact of time and treatment on rotarod performance. The analysis showed a statistically significant interaction between time and treatment (F_{(12, 176)} = 119.6, P < 0.001). Further examination of simple main effects indicated that time significantly influenced rotarod performance (F_{(3.442, 151.5)} = 182.9, P < 0.001), as did the treatment (F_{(3, 44)} = 486.0, P < 0.001). Additionally, Dunnett’s multiple comparisons test revealed that the sham group’s rotarod performance was significantly higher than that of the LPS group from week 1 to week 4 (P < 0.001). Similarly, the rotarod performance of the ISO + LPS group was also significantly greater than that of the LPS group during the same period (P < 0.001).

Figure 2.

Assessment of motor functions across various treatment groups through rotarod, pole-climbing and beam-walking tests. (A) Time-course changes in rotarod tests: different patterns of motor behavior performance were observed in the rotarod tests. Asterisks indicate statistical significance between groups (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (B) AUC analysis was conducted for rotarod performance in each group, illustrating the total performance over 5 weeks. Asterisks indicate statistical significance between groups (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (C) Time-course changes in pole-climbing tests: different patterns of motor behavior performance were observed through the pole-climbing. Asterisks indicate statistical significance between groups (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (D) AUC analysis was conducted for pole-climbing performance in each group, illustrating the total performance over 5 weeks. Asterisks indicate statistical significance between groups (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (E) Beam-walking performance differences across groups: The bar chart illustrates the decline in performance in the LPS group before and after surgery. In contrast, the other three groups show improvement in beam-walking tests. Asterisks indicate statistical significance between groups (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (F) Statistical analysis was conducted for the percentage differences in beam-walking performance. Statistical significance between groups is indicated by asterisks (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group).

If we considered the performance of the mice from baseline to four weeks post-surgery as a whole and analyzed the area under curve (AUC) for each group using one-way ANOVA, we observe statistically significant differences among the groups (F_{(3, 44)} = 13.464, P < 0.001). Post-hoc Dunnett T3 tests revealed that the AUC of the LPS group was significantly lower than the sham group (P < 0.001), and there was no statistically significant difference between the ISO group and the sham group. Conversely, the AUC of the ISO + LPS group was significantly higher than the LPS group (P < 0.001), as depicted in Figure 2(B).

Pole-climbing tests evaluated motor function in rodents by measuring the time taken to descend. Following LPS injection, the average descent time increased from 4 to nearly 8 seconds in the LPS group after 1 week, contrasting with the sham group and ISO group’s consistent 4-second performance. Remarkably, the ISO + LPS group completed the test in an average of 4.7 seconds, as depicted in Figure 2(C). A two-way repeated measures ANOVA was performed to evaluate the effects of time and treatment on pole-climbing performance. The results indicated a statistically significant interaction between time and treatment (F_{(12, 200)} = 7.774, P < 0.001). Further analysis of simple main effects revealed that both time (F_{(2.839, 142.0)} = 39.43, P < 0.001) and treatment (F_{(3, 50)} = 19.24, P < 0.001) significantly affected pole-climbing performance. Additionally, Dunnett’s multiple comparisons test showed that the sham group’s pole-climbing performance was significantly higher than that of the LPS group from week 1 to week 4 (P < 0.05 to P < 0.001). Similarly, the ISO + LPS group’s performance was also significantly better than that of the LPS group during the same timeframe (P < 0.05 to P < 0.001).

If we converted the results of each group to AUC for one-way ANOVA analysis, the results revealed a significant difference among groups (F_(3,50) = 20.095, P < 0.001). Post-hoc Dunnett T3 tests indicated that the AUC of the LPS group was significantly higher than the sham group (P < 0.001), and there was no statistical significance between the ISO group and the sham group. Conversely, the AUC of the ISO + LPS group was significantly lower than the LPS group (P < 0.01), as depicted in Figure 2(D).

The beam-walking tests assessed motor ability and balance by measuring the time it took for mice to traverse a 100-cm long and 12-mm wide beam both before and four weeks after surgery. The Kruskal-Wallis test revealed significant differences in the final test results among the groups (x²₍₃₎ = 32.7, P < 0.001). Dunn’s multiple comparisons test showed that the LPS group exhibited a significantly increased traversal time in the final tests compared to the sham group (P < 0.01). In contrast, the other three groups demonstrated a reduction in traversal time, with a significant difference observed between the ISO + LPS group and the LPS group (P < 0.01), as shown in Figure 2(E). The percentage difference (final tests minus baseline, divided by baseline time) indicated behavioral performance changes for each group. The Kruskal-Wallis test indicated significant differences in percentage difference among groups (x²₍₃₎ = 25.16, P < 0.001). Dunn’s multiple comparisons test revealed significant differences between the sham and LPS groups (P < 0.001), as well as between the LPS and ISO + LPS groups (P < 0.01). There was no statistical significance between the ISO group and the sham group, as depicted in Figure 2(F). Overall, based on three different motor behavioral tests, isofraxidin appeared to alleviate LPS-induced Parkinsonism-like motor dysfunction.

Isofraxidin Enhances Glucose Utilization in Cells Within the Striatal Region Affected by LPS: An [¹⁸F]FDG PET Imaging Analysis

To explore the effects of LPS injection into the striatum and the impact of pre-treatment with isofraxidin, we employed [¹⁸F]FDG PET imaging to evaluate glucose uptake five weeks post-surgery in each group. This assessed in vivo striatal glucose metabolism, indicative of neuronal and synaptic activity. ²⁰ Different [¹⁸F]FDG uptake patterns were observed among groups. The LPS group showed lower [¹⁸F]FDG standardized uptake value (SUV) compared to the sham group, suggesting reduced glucose uptake and metabolism in striatal cells and neurons due to LPS (Figure 3(A)). Figure 3(B) depicted quantitative results of [¹⁸F]FDG radioactivity in regions of interest. A Kruskal-Wallis H test was conducted to compare glucose SUV differences among groups within the striatum, revealing significant differences, x²₍₃₎ = 11.399, P = 0.01. Dunn’s multiple comparisons test further showed statistical significance, with the sham group differing significantly from the LPS group (P < 0.01) and the LPS group differing significantly from the ISO + LPS group (P < 0.05). No statistically significant difference was found between the ISO group and the sham group.

Figure 3.

[¹⁸F]FDG PET imaging and standardized uptake values (SUV) across different treatment groups. (A) Coronal view of reconstructed [¹⁸F]FDG PET imaging: In the figure, areas marked with orange circles indicate the locations of the striatum that were analyzed. Warmer tones (red) represent stronger signal intensity, indicating higher uptake of [¹⁸F]FDG. Conversely, cooler tones (blue) indicate weaker signal intensity and lower uptake of [¹⁸F]FDG. (B) Average SUV of glucose in striatum and statistical analysis: In the context of the Kruskal-Wallis test, Dunn’s multiple comparisons test reveals significant distinctions (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group).

Isofraxidin Restores Dopaminergic Fiber Density in the Striatum Following LPS-Induced Reduction

Figure 4(A) visually depicts TH staining of the striatum in each group, revealing reduced TH-positive fiber density in the LPS group compared to the sham group. Quantification of TH-positive cells in 20 μm brain samples involved optical density (OD) ratio analysis: ((OD_striatum-OD_{corpus callosum, cc}) / OD_cc). One-way ANOVA of TH OD ratio showed a statistically significant difference among groups (F_(3,24) = 11.467, P < 0.001). Figure 4(B) illustrates that the LPS group exhibited a significantly lower TH OD ratio than the sham group (P < 0.001), but there was no statistically significant difference between the ISO group and the sham group. However, the ISO + LPS group showed a statistically significant restoration of TH OD ratio compared to the LPS group (P < 0.05).

Figure 4.

Iba-1, TNF-α, IL-1β and TH staining across different treatment groups. (A) Images of TH staining in striatum. (B) Statistical analysis of OD ratio of TH in striatum: Within the LPS group, a noteworthy observation emerges as the OD ratio exhibits significant reduction compared to the sham group. This suggests that the injection of LPS leads to a decrease in the density of TH within the striatum. When comparing the ISO + LPS group to the LPS group, the significant disparity between these two groups indicates that isofraxidin potentially offers a protective effect on TH within the striatum, mitigating the adverse impacts of LPS-induced inflammation in the CNS (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (C) The morphology and quantity of microglia: upon administering LPS into the striatum, a noticeable elevation in the number of microglia is observed within the LPS group, in contrast to the Sham group. Moreover, when comparing the ISO + LPS group to the LPS group, a reduction in microglial expression is evident. (D) Statistical analysis for area percent of Iba-1-positive neurons in striatum (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (E) Images of TNF-α staining in striatum. (F) Statistical analysis of OD of TNF-α in striatum (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group). (G) Images of IL-1β staining in striatum. (H) Statistical analysis of OD of IL-1β in striatum (^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 vs sham group; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001 vs LPS group).

Isofraxidin Inhibits Microglial Accumulation in the Striatum Following LPS-Induced Activation

Through Iba-1 staining, we quantified microglial quantity and observed their morphology. Representative brain sections of each group revealed distinct patterns, with an abundance of Iba-1-positive cells observed in the LPS group compared to others, indicating significant microglial accumulation post-LPS injection into the striatum (Figure 4(C)). One-way ANOVA for area percent of Iba-1-positive neurons showed a significant difference (F_(3,24) = 18.725, P < 0.001). The Bonferroni post hoc analysis revealed a markedly increased area percent of Iba-1-positive neurons in the LPS group compared to the sham group (P < 0.001), but there was no significant difference between the ISO group and the sham group. Conversely, the ISO + LPS group exhibited a significant reduction in the area percent of Iba-1-positive neurons compared to the LPS group (P < 0.01) (Figure 4(D)), suggesting isofraxidin’s inhibitory effect on microglial accumulation in the striatum following stimuli such as LPS injection.

Isofraxidin Reduces LPS-Induced TNF-α and IL-1β Expression in the Striatum, Mitigating Inflammatory Responses

Figure 4(E) displays IHC staining of TNF-α across groups, showing elevated TNF-α expression in the LPS group. One-way ANOVA on OD of the striatum revealed a significant difference (F_(3,24) = 45.995, P < 0.001), as depicted in Figure 4(F). Dunnett T3 post hoc analysis indicated a significant difference between the LPS and sham groups (P < 0.001), indicating a strong inflammatory response in the striatum post-LPS injection. Conversely, the ISO + LPS group exhibited a notable decrease in TNF-α expression compared to the LPS group (P < 0.05), suggesting that isofraxidin pre-treatment could ameliorate LPS-induced inflammation.

Figure 4(G) illustrates the results of IL-1β staining, mirroring the trend observed in TNF-α staining across groups. The heightened staining intensity in the LPS group indicated an intensified inflammatory response triggered by LPS exposure. In Figure 4(H), one-way ANOVA on OD of the striatum showed a significant difference among groups (F_(3,24) = 30.592, P < 01). Dunnett T3 post hoc analysis revealed a statistical significance between the LPS and the sham groups (P < 0.01). Conversely, a significantly lower OD value was observed in the ISO + LPS group compared to the LPS group (P < 0.05). In the staining results for TNF-α and IL-1β, there was no statistically significant difference between the ISO group and the sham group.

Discussion

In this study, we explored the protective effects of isofraxidin on LPS-induced PD in mice. Our results demonstrated that administering isofraxidin before LPS exposure notably enhanced motor function in various behavioral tests. Additionally, [¹⁸F]FDG PET imaging revealed that isofraxidin preserved glucose uptake in neurons and metabolic function in the striatum. TH staining showed isofraxidin’s ability to prevent the loss of dopaminergic neurons induced by LPS. Furthermore, isofraxidin inhibited microglial activation and accumulation in the striatum, as seen in Iba-1 staining. Lastly, TNF-α and IL-1β staining indicated a reduction in inflammatory responses with isofraxidin pre-treatment.

LPS, a component of Gram-negative bacteria, induces a strong inflammatory response when injected into the brain, especially in areas like the striatum or substantia nigra. This leads to selective dopaminergic neuron loss in the nigrostriatal pathway, a hallmark of PD. ⁷ Compared to other commonly used models, such as 6-hydroxydopamine (6-OHDA) and MPTP, the LPS model offers several advantages. It closely replicates PD-like neuroinflammation, characterized by microglial activation, the release of pro-inflammatory cytokines (e.g., TNF-α, IL-1β), upregulation of COX-2 gene expression, and the production of reactive oxygen species (ROS). ²¹ This chronic inflammatory environment also promotes the aggregation of α-synuclein, forming Lewy bodies, which are strongly linked to PD pathology. ²² In contrast, the 6-OHDA model primarily affects the nigrostriatal pathway and does not produce Lewy body-like inclusions.^23,24 Moreover, 6-OHDA cannot cross the BBB, ²⁵ requiring direct intracranial administration into regions such as the cistern, ventricle, or brain parenchyma. The MPTP model, while effective and commonly used, can be limited by species differences ²⁶ and lacks consistent Lewy body formation. ²⁷ The LPS model, by contrast, provides a more comprehensive representation of both inflammation and dopaminergic neuron degeneration. Additionally, humans are frequently exposed to LPS through environmental factors like air pollution and particulate matter, making this model relevant for studying the potential environmental triggers of PD. ²⁸ Based on these advantages, the LPS model was chosen for this study to explore therapeutic interventions targeting both neuroinflammation and neuronal loss in PD.

While isofraxidin has been reported for its neuroprotective effects, research specifically addressing its impact on motor impairments and microglia in PD remains limited. Furthermore, most studies have primarily focused on plant extracts, such as Acanthopanax senticosus, a common source of isofraxidin, rather than examining the pure compound itself. In our study, by using the pure compound, we can more accurately assess its therapeutic potential. In spite of the fact that Acanthopanax senticosus extracts have shown protective effects in MPTP PD models, ¹⁶ our study employs an LPS-induced PD model that is more relevant to human pathology. Additionally, most research on isofraxidin has focused on peripheral diseases, with few studies examining its impact on CNS disorders. Our findings demonstrate that isofraxidin can be orally absorbed, cross the blood-brain barrier, and reduce LPS-induced inflammation, suggesting its clinical value as an oral treatment option for PD. This highlights the distinct contributions of our research, even in the context of prior studies.

Our research found that mice injected with LPS in the striatum exhibited a significant decline in behavioral performance. However, the motor dysfunction in the ISO + LPS group was markedly alleviated (Figure 2(A), (C), and (E)). Our IHC staining results suggest that isofraxidin may mitigate LPS-induced damage in the striatum due to its anti-inflammatory effects, as indicated by significantly reduced levels of TNF-α and IL-1β (Figures 4(E) and (G)). In addition to these pro-inflammatory factors, isofraxidin has been shown to improve outcomes in other contexts. For instance, it has been found to alleviate myocardial infarction through NLRP3 inflammasome inhibition. ²⁹ Furthermore, isofraxidin has demonstrated efficacy in improving memory deficits in scopolamine-treated mice by suppressing oxidative stress and inflammatory responses, as well as modulating the BDNF-CREB-ERK signaling pathway. ³⁰ Another study revealed that isofraxidin alleviates depressive-like behaviors resulting from chronic unpredictable mild stress, linking these effects to reduced inflammation and regulation of the hypothalamic-pituitary-adrenal (HPA) axis. ³¹ These findings suggest that the behavioral improvements observed in our study may also be associated with isofraxidin’s capacity to address mood-related deficits and upregulate BDNF-CREB-ERK signaling, both of which support neuroplasticity and cognitive function. This highlights isofraxidin’s potential for improving overall behavioral performance in contexts of neuroinflammation and stress. Additionally, the persistent motor deficits observed in our study, which are also noted in other PD studies,^14,16 underscore the lasting impact of neuroinflammation on motor performance. This has important implications for understanding the progression of PD and highlights the need for early therapeutic interventions to mitigate such deficits.

[¹⁸F]FDG PET imaging is primarily used to detect abnormal glucose uptake and metabolism. ³² In animal models of PD, in vivo [¹⁸F]FDG PET imaging has been employed to quantify alterations in regional glucose utilization, assessing metabolic changes.^33,34 In our study, we triggered microglia-induced inflammation by directly injecting LPS into the striatum, aiming to cause subsequent damage. Five weeks post-surgery, during the chronic inflammation phase, [¹⁸F]FDG PET imaging revealed a significant decrease in glucose uptake in the striatum of the LPS group compared to the sham group. We believe that the cellular and neuronal death in the striatum during the prolonged inflammatory response, leads to a significant reduction in glucose utilization in that area. With the treatment of isofraxidin, however, glucose utilization was effectively restored. A separate study on Alzheimer’s disease indicated that, besides neuronal death, neuritic atrophy and the loss of synaptic connections are major factors driving the progression of neurodegenerative disorders. The researchers found that extracts derived from Siberian ginseng (Eleutherococcus senticosus rhizome) displayed neuroprotective effects, promoting neurite regrowth and the restoration of synaptic networks in rat cortical neurons exposed to amyloid β (Aβ)(25-35) damage. Through a comprehensive analysis of the extract’s components, the active compounds identified—eleutheroside B, eleutheroside E, and isofraxidin—were shown to provide significant protection against the axonal and dendritic atrophy caused by Aβ(25-35). ³⁵ These findings suggest that compounds like isofraxidin may play a role in protecting against neuronal damage, potentially offering therapeutic avenues for neurodegenerative conditions such as PD, where both neuronal death and inflammation are common contributors. However, contrary findings from a study on LPS-induced lung inflammation suggested increased [¹⁸F]FDG uptake in the lungs during acute inflammation, likely due to elevated energy consumption from pulmonary neutrophil accumulation. ³⁶ Therefore, timing of [¹⁸F]FDG PET imaging may influence outcomes, emphasizing the importance of considering different stages of inflammation.

In PD, the degeneration of DA neurons in the SNpc leads to the loss of their projections to the striatum, while activated microglia accumulate in the affected areas. ³⁷ Iba-1 is a specific marker expressed in macrophages and microglia, and its expression is upregulated upon the activation of these cells. ³⁸ Quantifying Iba-1 expression provides a reliable method to assess microglial activity in the CNS. In the sham group, microglia predominantly exhibit a ramified morphology, characterized by elongated branches and a small cell body, indicating a resting or surveilling state as they monitor brain tissue for damage or pathogens. In contrast, the LPS group shows significant changes, with microglia adopting an amoeboid shape, reflecting heightened secretion of inflammatory cytokines and neurotrophic factors.^39,40 In addition to this morphological transformation, we also observed a notable accumulation of microglia in the striatum of the LPS group. In contrast, isofraxidin effectively reduces this accumulation and attenuates the associated inflammatory response, highlighting its potential as a therapeutic agent in reducing neuroinflammation. Moreover, research suggests that isofraxidin may promote the polarization of macrophages towards anti-inflammatory M2a and M2b phenotypes, which could further contribute to its immunoregulatory activities. ⁴¹

Lastly, our results show a substantial reduction in the OD ratio of TH within the striatum of mice in the LPS group, and isofraxidin significantly mitigates the loss of TH (Figure 4(B)). This finding emphasizes the potential of isofraxidin to counteract the detrimental effects of LPS on dopaminergic neurons in the striatum. Furthermore, isofraxidin is recognized as an inhibitor of monoamine oxidase B (MAO-B), ⁴² which may contribute to an increase in dopamine concentrations. By inhibiting MAO-B, isofraxidin could help maintain higher levels of dopamine, potentially leading to improved behavioral performance. The relationship between dopamine concentration and behavior suggests that isofraxidin’s neuroprotective effects may not only stem from its ability to preserve dopaminergic neurons but also from its capacity to enhance dopamine availability, ultimately benefiting motor and cognitive functions affected by neuroinflammation.

Although isofraxidin has been shown to cross the BBB in both our results and in other research focused on its use for central nervous system disorders such as Alzheimer’s disease and PD, its precise metabolism, distribution, and concentration within the CNS remain unclear. This lack of detailed information may impact our understanding of its therapeutic efficacy, specifically in targeting neuroinflammation and protecting dopaminergic neurons in the CNS. Further research is needed to investigate isofraxidin’s localization, metabolic pathways, and achievable concentrations within central neural tissue, as these factors could significantly influence its effectiveness in alleviating PD symptoms.

Conclusion

In conclusion, our study provides compelling evidence that isofraxidin exerts neuroprotective effects in a mouse model of PD induced by LPS. Isofraxidin not only improved motor functions and preserved glucose metabolism in the striatum but also effectively reduced microglial activation and the subsequent inflammatory response. These findings highlight the compound’s ability to mitigate the loss of dopaminergic neurons, as indicated by preserved TH levels and decreased inflammatory cytokines. Furthermore, the relevance of the LPS model in mimicking human PD pathology emphasizes the potential of isofraxidin as a promising oral treatment option. Given the persistent nature of neuroinflammation in PD, early intervention with isofraxidin may play a crucial role in slowing disease progression and enhancing neuronal resilience. Further research is needed to explore its clinical applications and the underlying mechanisms contributing to its protective effects.

Supplemental Material

Supplemental Material - Neuroprotective potential of isofraxidin: Alleviating parkinsonian symptoms, inflammation and microglial activation

Supplemental Material for Neuroprotective potential of isofraxidin: Alleviating parkinsonian symptoms, inflammation and microglial activation by Tin-An Wang, Shiao-Yun Li, Li-Yun Fann, I-Hsun Li, Tsung-Ta Liu, Hao-Yuan Hung, Chieh-Wen Chang, Chih-Chien Cheng, Ying-Che Huang, Pei-Yeh Yu, and Jui-Hu Shih in Journal of Central Nervous System Disease.