Skip to main content
Download PDF

CEST imaging combined with 1H-MRS reveal the neuroprotective effects of riluzole by improving neurotransmitter imbalances in Alzheimer’s disease mice

Alzheimer's Research & Therapy volume 17, Article number: 20 (2025) Cite this article

Abstract

Background

The imbalance of glutamate (Glu) and gamma-aminobutyric acid (GABA) neurotransmitter system plays a crucial role in the pathogenesis of Alzheimer’s disease (AD). Riluzole is a Glu modulator originally approved for amyotrophic lateral sclerosis that has shown potential neuroprotective effects in various neurodegenerative disorders. However, whether riluzole can improve Glu and GABA homeostasis in AD brain and its related mechanism of action remain unknown. This study utilized chemical exchange saturation transfer (CEST) imaging combined with proton magnetic resonance spectroscopy (1H-MRS) to monitor the dynamic changes of Glu and GABA in riluzole-treated AD mice, aiming to evaluate the efficacy and mechanism of riluzole in AD treatment.

Methods

GluCEST, GABACEST and 1H-MRS were used to longitudinally monitor Glu and GABA levels in 3xTg AD mice treated with riluzole (12.5 mg/kg/day) or vehicle for 20 weeks. Magnetic resonance measurements were performed at baseline, 6, 12, and 20 weeks post-treatment. Cognitive performance was assessed using the Morris Water Maze (MWM) at baseline, 10, and 20 weeks. At the study endpoint, immunohistochemistry, Nissl staining, and Western blot were used to evaluate the brain pathology, neuronal survival, and protein expression.

Results

GluCEST, GABACEST and 1H-MRS consistently revealed higher levels of Glu and GABA in the brain of riluzole-treated AD mice compared to untreated controls, which were associated with improvements in spatial learning and memory. The cognitive improvements significantly correlated with the increased GluCEST signals and Glu levels. Immunohistochemistry and Nissl staining demonstrated that riluzole treatment reduced amyloid-beta (Aβ) deposition, tau hyperphosphorylation, GFAP-positive astrocyte activation, and prevented neuronal loss. Moreover, riluzole upregulated the expression of excitatory amino acid transporter 2 (EAAT2), glutamic acid decarboxylase 65/67 (GAD65/67), and glutamine synthetase (GS), suggesting enhanced neurotransmitter metabolism.

Conclusions

CEST imaging combined with 1H-MRS demonstrated the effectiveness of riluzole in modulating Glu- and GABA-related changes and improving cognitive function in 3xTg AD mice, potentially through regulating key proteins involved in neurotransmitter metabolism. These findings suggest riluzole as a therapeutic agent for Alzheimer’s disease and highlight the utility of multimodal MR imaging in monitoring treatment response and exploring disease mechanisms.

Introduction

Alzheimer’s disease (AD), a progressive neurodegenerative disorder, is characterized by a gradual decline in cognitive function and memory. Studies showed a significant increase in the incidence and prevalence of AD from 1990 to 2019, by 147.95% and 160.84%, respectively [1,2,3]. By 2050, an estimated 152 million people will suffer from AD and other dementias, causing substantial distress to patients and placing a significant financial and caregiving burden on families and society [4]. The annual economic impact varies depending on disease severity, ranging from $468.28 for mild cases to $171,283.80 for severe cases [5].

The etiology of AD remains unclear, but studies have shown that the primary neuropathological hallmarks are the deposition of extracellular amyloid-beta (Aβ) plaques and intracellular neurofibrillary tangles composed of hyperphosphorylated tau proteins. These pathological alterations cause neuronal damage and synaptic loss, leading to cognitive impairments and memory deficits [6, 7]. Additionally, disruption of neurotransmitter systems, particularly the imbalance between glutamate (Glu) and gamma-aminobutyric acid (GABA), also plays a significant role in the progression of AD. Glu, a major excitatory neurotransmitter, has been found to be significantly reduced in the cortex and hippocampus in AD patients, highlighting a potential mechanism for the cognitive and memory deficits [6]. Similarly, the GABAergic system, the main inhibitory system in the mammalian brain, has been found to be disrupted in various brain disorders, including AD, with lower GABA levels in AD patients compared to healthy controls [8]. Despite extensive research into the pathogenesis of AD, there is currently a lack of treatments that can significantly improve patient prognosis. Current drugs approved by the Food and Drug Administration (FDA) mainly focus on managing symptoms and slowing disease progression, but do not significantly alter the course of the disease. Some newer monoclonal antibody therapies have shown initial disease-modifying effects, but overall efficacy remains limited [9,10,11]. Therefore, there is an urgent need to develop new therapeutic strategies and effective monitoring methods to improve the quality of life and prognosis of AD patients.

Riluzole was originally used to treat amyotrophic lateral sclerosis (ALS) by acting as a Glu modulator to inhibit excessive release of Glu, thereby protecting neurons from excitotoxic damage [12]. Recent investigations have elucidated its potential efficacy in treating AD. Previous studies have demonstrated that in TauP301L and AβPP/PS1 AD mouse models, treatment with riluzole attenuated glutamate release and glutamate uptake in the hippocampus, which is associated with enhanced cognitive performance [13, 14]. Furthermore, in a recently published article, the riluzole prodrug troriluzole has been shown to suppress expression of glutamate transporter 1 (VGlut1) in presynaptic vesicules of 3xTg-AD mice, inhibiting basal and stimulus-induced glutamate release and subsequent cognitive improvement [15]. In addition to enhancing glutamatergic neurotransmission, riluzole has also been shown to regulate the glutamatergic/GABAergic balance in the dentate gyrus of the hippocampal, thereby significantly affecting synaptic plasticity and restoring cognitive functions [14, 16, 17]. In addition, riluzole also has been shown to be effective in reducing the deposition of beta-amyloid plaques and tau protein tangles [18, 19]. Despite the encouraging results of riluzole in AD-related animal models, its long-term effects and the dynamic changes of neurotransmitter metabolism during treatment still require further investigation.

Proton magnetic resonance spectroscopy (1H-MRS) and chemical exchange saturation transfer (CEST) imaging are advanced techniques that can provide detailed information on specific metabolite levels in the brain. 1H-MRS studies have documented reductions in N-acetylaspartate (NAA) and Creatine (Cr) in AD patients, indicating neuronal damage and mitochondrial dysfunction [20]. However, the measurement of metabolites related to neurotransmitters, such as Glu and GABA, have shown variability [21,22,23]. CEST imaging, especially GluCEST and GABACEST, has shown promise in assessing the dynamics changes of Glu and GABA in various brain disorders [24,25,26,27,28,29,30]. However, the role of CEST imaging in the long-term assessments of AD is still needed to be explored.

In this study, GluCEST and GABACEST combined with 1H-MRS were used to dynamically monitor the changes of Glu and GABA in 3xTg AD mice treated with riluzole. The therapeutic effects and mechanism of riluzole were evaluated by combining in vivo MR-based metabolic findings with behavioral and pathological results. Our results indicated that riluzole may exert its neuroprotective effects by modulating the levels of Glu and GABA in the AD mouse model, and CEST imaging combined with 1H-MRS can be used as a non-invasive method for long-term monitoring of drug response and to explore neurotransmitter dysfunction in the pathogenesis of AD.

Methods

Animal preparation

The 3xTg AD female mice (7 months old, from Jiangsu Ailinfei Biotechnology Co., Ltd) and age-matched C57BL/6J mice (from Zhuhai Baishitong Biotechnology Co., Ltd) were used. All animal experiments were approved by the Ethics Committee of Shantou University Medical College (Approval ID: SUMC2022-597) and conducted in accordance with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. At 7 months of age, 3xTg AD mice (n = 9) and C57BL/6J mice (n = 8) were randomly assigned to receive riluzole (12.5 mg/kg/day) in drinking water. Another group of 3xTg AD (n = 9) and C57BL/6J (n = 8) mice were given blank water without riluzole as a control [13, 14]. Twenty-four hours after the final behavioral testing, the mice were anesthetized, and brain tissue was collected for histological characterization and Western blotting. A timeline of the experimental design was presented in Fig. 1.

Fig. 1
figure 1

Experimental design diagram. Four groups of mice were treated as follows: Riluzole-treated 3xTg AD mice (AD_R, n = 9), untreated 3xTg AD mice (AD_C, n = 9), Riluzole-treated C57BL/6J mice (C57_R, n = 8), and untreated C57BL/6J mice (C57_C, n = 8). Starting from 7 months of age (7 M), mice in AD_R group and C57_R group were administered riluzole in drinking water at a dose of 12.5 mg/kg. This treatment was lasted for 20 weeks, and MR-based metabolic measurement was performed at baseline (7 M), 6 weeks (8.5 M), 12 weeks (10 M), and 20 weeks (12 M) after treatment. Mice also underwent cognitive and behavioral evaluation using Morris water maze (MWM) at baseline (7 M) and at 10 weeks (9.5 M) and 20 weeks (12 M). MR-based metabolic evaluation and histological analysis were performed at the end of treatment (12 M)

Full size image

MRI preparation

Prior to MR scanning, mice were anesthetized using 3.0% isoflurane at an oxygen- induced flow rate of 0.5 L/min, then maintained anesthesia with 1.5 − 1.8% isoflurane. Respiratory rates were continuously monitored with a respiratory probe (model 1030, SAII) during MR scanning.

MRI acquisition

MR scans were performed on a 7.0 T animal MR system (Agilent Technologies, CA, USA) with a 72 mm volume coil for transmission and a brain surface coil (Time Medical Technologies, China) for signal reception. 3xTg AD mice and C57BL/6J mice were scanned at 7 months of age (baseline), 6 weeks, 12 weeks, and 20 weeks after riluzole treatment. T2-weighted imaging was performed with the following parameters: repetition time (TR) = 2000 ms, echo time (TE) = 24.48 ms, slice thickness = 2 mm, number of slices = 7, gap = 0.02 mm, field of view = 22 × 22 mm, data matrix = 128 × 128, and scanning time = 1 min 8 s. Prior to CEST imaging, B0 field inhomogeneities were corrected using the Water Saturation Shift Referencing (WASSR) method [31], and B1 field calibration was performed using the double flip angle approach [32].

Chemical exchange saturation transfer (CEST) imaging

A gradient recalled echo (GRE) sequence with a continuous-wave pre-saturation pulse (B1 = 4 µT, pulse width = 2 s) was used to perform Glu-weighted-CEST (GluCEST) and GABA-weighted-CEST (GABACEST) imaging with the following parameters: repetition time = 5000 ms, echo time = 14 ms, slice thickness = 2 mm, number of slices = 1, kzero = 32, shot = 1, field of view = 22 × 22 mm, data matrix = 64 × 64. The saturation frequency ranged from − 5 to 5 ppm in increments of 0.1 ppm, with a reference offset (non-saturated) image at − 100 ppm, and the scanning time was 14 min 15 s. To validate the reliability and specificity of the chosen saturation frequencies for detecting changes in GABA and glutamate levels, we performed a phantom study using samples containing varying concentrations of these metabolites. The detailed methods and results of the phantom study are provided in the supplementary materials (Figure S1).

Proton magnetic resonance spectroscopy (1H-MRS)

1H-MRS was employed to acquire the metabolic profiles. The volumes of interest (VOIs) were placed in the right hippocampus (2.5 mm × 2.5 mm × 2.5 mm) and frontal cortex (1.2 mm × 2.5 mm × 2.5 mm) based on T2-weighted images. 2D shimming was performed automatically to optimize the first- and second-order shim values in the VOI. The shimming quality was visually inspected in the non-water suppressed spectrum with a water peak linewidth lower than 20 Hz. In vivo localized spectra were acquired using a point resolved spectroscopy sequence (PRESS) and variable angle spin echo with optimized relaxation delays (VAPOR) were used for water suppression. The acquisition parameters were as follows: TR = 3000 ms, TE = 15.92 ms, Bandwidth = 50,000 Hz, Number of Averages = 256, acquisition time = 13 min and 3 s.

Data analysis

Matlab 2022 (MathWorks, Natick, MA) was used for post-processing of the CEST images. Regions of interest (ROIs) including total brain, bilateral hippocampi and cortex were manually delineated based on T2-weighted images. Following B0 correction, Z-spectra were generated by calculating the ratio of the signal obtained at various irradiation frequency offsets to the unsaturated water protons (S0) signal. The GluCEST and GABACEST maps were generated by the relative changes in percentage units as follows: CESTR (%) = 100 × [(S–ω – S+ω)/S–ω], where CESTR was normalized to the signal losses by direct water saturation and magnetization transfer at –ω, and S–ω and S+ω were B0 corrected signals for the water resonance of GluCEST at − 3.0 and + 3.0 ppm, and of GABACEST at − 2.75 and + 2.75 ppm, respectively.

The spectral quality was assessed by measuring the signal-to-noise ratio (SNR) and linewidth of the unsuppressed water peak. Spectra with SNR > 10 and linewidth < 10 Hz were considered to be of acceptable quality for further analysis. Representative spectra from hippocampus and cortex are provided in Fig. 2A. LCModel (version 6.3–1 L, Stephen Provencher, Oakville, ON, Canada) was used to quantify metabolite concentrations. The basis set included spectra for alanine (Ala), aspartate (Asp), creatine (Cr), phosphocreatine (PCr), γ-aminobutyric acid (GABA), glucose (Glc), glutamate (Glu), glutamine (Gln), glycerophosphocholine (GPC), phosphocholine (PCh), myo-Inositol (mI), lactate (Lac), N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), scyllo-inositol (Ins), and taurine (Tau). Only metabolites with Cramer-Rao Lower Bounds (CRLB) ≤ 25% were included in the analysis. Overlapping peaks were handled by the LCModel software which uses a linear combination of individual spectra to estimate the contribution of each metabolite. The spectra range between 0.2 and 4.2 ppm were post-processed, and 8 major metabolites including NAA, Cr + PCr, Glu, Gln, GABA, tCho, Tau and Ins were quantified by LCModel.

Fig. 2
figure 2

Dynamic monitoring of metabolites by 1H-MRS in 3xTg AD mice treated with riluzole. (A) T2-weighted coronal MRI images showed regions of interest (ROIs) in the right hippocampus (upper left) and cortex (lower left), with corresponding 1H-MRS spectra illustrating metabolic profiles. (B) Changes in metabolite concentrations in hippocampus and cortex for four groups at baseline, 6, 12, and 20 weeks post-treatment. (C) Comparative analysis across four groups highlighted metabolite concentrations in hippocampus and cortex at 20 weeks post-treatment. Data were presented as mean ± SD. Statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Morris water maze training and probe challenge

All 3xTg AD mice and C57BL/6J mice underwent cognitive assessments using the MWM spatial learning and memory recall paradigm at baseline (prior to treatment), and then at 10 and 20 weeks after treatment. The MWM protocol consisted of a 1-day visible platform test followed by 4 days of hidden platform training. On day 1, the mice were tested four 120-second trials on a visible platform. On days 2–5, the mice were tested four 60-second trials daily on a submerged platform, with 20-minute intervals between each trial. If the mouse found the submerged platform within 60 s, it was allowed to stay on the platform for 20 s. If the mouse did not find the platform, it was directed to the platform and allowed to stay for 20 s.

Twenty-four hours after the final training trial, the platform was removed and the mice were allowed to swim freely for 60 s to test their memory retention in a probe trial. The performance was recorded and analyzed using a video analysis system provided by Zhongshidi Chuang Science and Technology Development Co., Ltd. The metrics included the escape latency during the learning trials, as well as the occupancy time and distance within the target quadrant during the probe trial. The percentage of time spent in the target quadrant was calculated as: target quadrant time / total testing time × 100%. Similarly, the percentage of path length in the target quadrant was calculated as: target quadrant path length / total path length × 100%.

Immunohistochemistry

For histological characterization, the brain tissue was fixed with 4% paraformaldehyde (4°C, 48 hours), dehydrated, and embedded in paraffin. The brain tissues of AD_R and AD_C mice (n = 4), and C57_R and C57_C mice (n = 3) were continuously sliced 10-µm in the coronal plane for immunohistochemical staining. The primary antibodies used were rabbit anti- tau phosphorylated at serine 404 (Tau pS404) (1:100; ab92676, Abcam, Cambridge, MA, United States), rabbit anti- glial fibrillary acidic protein (GFAP) (1:100; ab7260, Abcam), and rabbit anti-β-Amyloid (1:100; 715800, Thermo Scientific). The secondary antibody used was goat anti-rabbit IgG (H + L) (1:1000; ab6721, Abcam). DAB solution was used for immunochemistry. For each mouse brain, three non-consecutive and non-overlapping fields of view were randomly chosen in the hippocampus and parietal cortex at 20x magnification. The number of immunopositive areas or cells was quantified using the ImageJ software and expressed as an average for each high-power field of view (HPF).”

Nissl staining

Nissl staining was employed for further assessment of neurological damage. Paraffin sections were dewaxed, hydrated, stained with 1% toluidine blue, differentiated, dehydrated, and sealed for microscopic observation. Three consecutive but non-overlapping fields of view (20 times magnification) were randomly selected in the hippocampus and parietal cortex of each mouse brain tissue. The neurons were counted using the ImageJ software and represented as the average of each high-power field of view (HPF).

Western blotting

Proteins were extracted from hippocampus and cortex using radioimmunoprecipitation assay (RIPA) buffer containing protease inhibitors. The protein concentration was determined using the bicinchoninic acid (BCA) reagent. Equal amount of protein was then subjected to polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Temecula, CA). The PVDF membrane was blocked overnight at 4 °C in 5% bovine serum albumin (BSA) in Tris-buffered saline with 0.1% Tween 20 (TBST), then incubated with primary antibodies, including excitatory amino acid transporter 2 (EAAT2) (1:1,000; ab205248, Abcam), glutamic acid decarboxylase 65/67 (GAD65/67) (1:1,000; ab183999, Abcam), glutamine synthetase (GS) (1:1,000; ab197024, Abcam), Aβ (1:1,000; 715800, Thermo Scientific), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:10,000; ab181602, Abcam) in TBST at room temperature for 2 h. After incubation with horseradish peroxidase (HRP)-conjugated secondary antibody (1:4,000), bands were visualized using enhanced chemiluminescence (ECL) and quantified by densitometry using ImageJ software. The relative intensities of each sample were normalized to GAPDH. Detailed information on the antibodies used in this study is provided in Supplementary Table S1.

Statistical analyses

Data was analyzed using GraphPad Prism (version 9.5.1) and SPSS statistical software (version 25). The Shapiro-Wilk test was employed to assess the normality of all quantitative data. Group comparisons were conducted using one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparisons test to control for Type I error inflation in the context of multiple pairwise comparisons. For CEST imaging 1H-MRS, and Morris Water Maze data, two-way ANOVA was used, followed by Tukey’s post hoc test for multiple comparisons to maintain a conservative approach to familywise error rate (FWER) control. Correlation analyses were conducted using Pearson’s correlation coefficient. Statistical significance was set at p < 0.05 (two-tailed) for all analyses, with this threshold determined in the context of Tukey’s multiple comparisons test. Data were presented as mean ± standard deviation.

Results

Dynamic CEST imaging of 3xTg AD mice treated with riluzole

GluCEST and GABACEST imaging of the whole brain, hippocampus, and cortex in AD_R, AD_C, C57_R, and C57_C groups at baseline, 6, 12, and 20 weeks post-treatment showed gradually elevated signals in the AD_R group (Fig. 3A and B). Two-way ANOVA revealed significant main effects of group and time × group interaction for GluCEST signals in the hippocampus (group: p < 0.0001; interaction: p = 0.002) and cortex (group: p < 0.0001; interaction: p = 0.003). Post hoc comparisons were conducted using Tukey’s multiple comparisons test. For GABACEST signals, group and time × group interaction were significant in the hippocampus (group: p < 0.0001; interaction: p = 0.006), while in the cortex, only the main effect of time (p = 0.02) and group (p < 0.0001) was significant.

Fig. 3
figure 3

Dynamic GluCEST and GABACEST evaluation of 3xTg AD mice treated with riluzole. (A, B) GluCEST and GABACEST images of the whole brain, hippocampus, and cortex in AD_R, AD_C, C57_R, and C57_C groups at baseline, 6, 12 and 20 weeks post-treatment. (C, D) Quantitative analysis of GluCEST and GABACEST signals in the hippocampus and cortex over time. (E, F) Comparative analysis of GluCEST and GABACEST signals in the hippocampus and cortex across four groups at 20 weeks post treatment. Data were presented as mean ± SD. Statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Quantitative analysis of GluCEST (Fig. 3C) and GABACEST (Fig. 3D) signals in the hippocampus and cortex revealed significant differences among four groups. Both GluCEST and GABACEST signals were significantly increased in AD_R group, suggesting that riluzole had neuroprotective effects. Specifically, GluCEST signals in the hippocampus increased from 0.154 ± 0.004 at baseline to 0.186 ± 0.005 at week 20. with a similar pattern in the cortex (baseline: 0.164 ± 0.004; 20 weeks: 0.194 ± 0.006). The GABACEST of AD_R groups also exhibited an upward trend. In the hippocampus, GABACEST increased from 0.127 ± 0.004 at baseline to 0.142 ± 0.004 at 12 week, and the increase was statistically significant (p = 0.050). Similarly, in the cortex, GABACEST rose from 0.131 ± 0.004 at baseline to 0.151 ± 0.005 at 12 week (p = 0.023). However, changes in GABACEST signals at 20 week did not reach statistical significance compared to baseline. The AD_C group showed a downward trend, while the C57BL/6J group remained stable.

At 20 weeks post-treatment, GluCEST and GABACEST signals in the AD_R group were comparable to those in the C57_R and C57_C groups (Fig. 3E and F; Supplementary Table S2, Figure S3), suggesting that riluzole treatment can effectively maintain Glu/GABA homeostasis in 3xTg AD mice. The AD_C group exhibited significantly reduced signals in both brain regions. Tukey’s multiple comparison test revealed that the GluCEST signals in hippocampus (p < 0.01) and cortex (p < 0.05) in AD_C group were significantly lower than those in the AD_R group and both control group (p < 0.05). Similarly, GABACEST signals in hippocampus and cortex in AD_C group were significantly lower than those in both control groups (p < 0.05 for all comparisons; Fig. 3F). Interestingly, while the AD_R group demonstrated a trend towards normalized GABA levels, the differences between the AD_R group and either the AD_C or control groups did not reach statistical significance. The significantly reduced GluCEST and GABACEST signals in the untreated AD_C group indicated an increased Glu/GABA homeostatic imbalance in untreated 3xTg AD mice.

Dynamic monitoring of metabolites by 1H-MRS in 3xTg AD mice treated with riluzole

Representative 1H-MRS spectra of the hippocampus and frontal cortex were presented in Fig. 2A. Two-way ANOVA revealed that group and time × group interaction had significant effects on the concentrations of Glu, GABA, and Gln in the hippocampus (Glu: group, p = 0.003; interaction, p = 0.025; GABA: group, p = 0.011; interaction, p = 0.002; Gln: group, p = 0.019; interaction, p = 0.014), while time had significant effects on Glu and Gln (Glu: p = 0.004; Gln: p = 0.0003) but not on GABA. Further analyses were conducted using Tukey’s multiple comparisons test. In the cortex, Glu concentration was affected by time (p = 0.032) and group (p < 0.0001), and GABA concentration was affected by time (p = 0.018), group (p = 0.003) and their interaction (p = 0.003). Cortical Gln concentration was only affected by time (p = 0.002).

Longitudinal measurements of major metabolite concentrations in the hippocampus and cortex were shown in Fig. 2B. In the riluzole-treated group (AD_R), Tukey’s post hoc analysis revealed significant increases in the levels of Glu (p = 0.013), GABA (p = 0.017), Gln (p = 0.036), NAA (p = 0.039), Cho (p = 0.012), and Cr (p = 0.034) at 20 weeks post-treatment compared to baseline in the hippocampus. Similarly, in the cortex, concentrations of Glu, Gln, NAA, Cho, and Cr were significantly higher in the AD_R group at 20 weeks compared to baseline (all p < 0.05). Glu and GABA concentrations in the AD_C group gradually decreased, while metabolic levels in the control group also remained relatively stable.

At 20 weeks post-treatment, metabolite concentrations were compared between groups (Fig. 2C). In the hippocampus, Glu levels in AD_C group were significantly reduced compared to AD_R (p = 0.01), C57_R (p = 0.011), and C57_C (p = 0.001) groups, with similar reductions observed for Gln, NAA, Cho, and Cr. In the cortex, Glu levels were lower in the AD_C group compared to the AD_R (p = 0.027), C57_R (p = 0.031) and C57_C (p = 0.042) groups. Cortical GABA levels in the AD_C group were also significantly decreased compared to the C57_R (p = 0.042) and C57_C (p = 0.013) groups. The levels of Cr also showed significantly lower in the AD_C group compared with those in other groups (p < 0.05). These results suggested that riluzole treatment maintained metabolite levels within normal range in 3xTg AD mice, while untreated AD mice had significant metabolic imbalance.

Riluzole improved cognition and learning abilities in 3xTg AD mice

MWM was used to assess riluzole’s effect on cognition and learning in 3xTg AD mice. In the spatial learning test (Fig. 4A upper panels, 4B), the AD_C group took significantly longer time to locate the hidden platform, compared to the other groups. Two-way ANOVA revealed a significant main effect of Group (p < 0.0001), with Tukey’s post-hoc test showing prolonged latencies in AD_C group compared to AD_R group (p = 0.0004), C57_R and C57_C groups (both p < 0.0001) at 20 weeks of treatment. Although the escape latency decreased in each group during the training period, the difference within the groups did not reach statistical significance.

Fig. 4
figure 4

Riluzole improved cognition and learning abilities in 3xTg AD mice. (A) Representative trajectories of the AD_R, AD_C, C57_R and C57_C groups in Morris Water Maze at 20 weeks post-treatment. The spatial learning tests (upper panels) and probe test (lower panels) were shown, and the AD_C group showed longer search times and more convoluted swimming paths in the target quadrant. (B) Spatial learning test conducted at 20 weeks post-treatment (days 2–5). The latencies of searching for the hidden platform were significantly longer in the AD_C group compared to the other three groups. The capped lines indicated significant differences between treatment groups. (C-D) Probe tests performed at baseline, 10 and 20 weeks post-treatment. At 10 and 20 weeks, the AD_C group showed significantly lower Target Quadrant Occupancy, measured as percentage of time (C) and pathlength (D), compared to the C57_R and C57_C groups. Data were presented as mean ± SD. Statistical significance was indicated by *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Full size image

During the probe test (Fig. 4A lower panels, Fig. 4C and D), the AD_C group showed reduced proficiency in the target quadrant time and distance. Two-way ANOVA showed significant main effects of Group (p = 0.004) and Time (p = 0.003). Tukey’s post hoc analysis showed that at 10 and 20 weeks post-treatment, both the time spent and the distance traveled in the target quadrant decreased significantly in the AD_C group compared to the C57_R group (10 weeks: time ratios, p = 0.004; distance ratios, p = 0.019; 20 weeks: time ratios, p = 0.038, distance ratios, p = 0.013) and C57_C group (10 weeks: time ratios, p = 0.019, distance ratios, p = 0.014, 20 weeks: time ratios, p = 0.014, distance ratios, p = 0.008). In contrast, no significant difference was observed between AD_R group and control groups at 20 weeks.

Correlations analysis among CEST/MRS measurements of Glu/GABA and cognition after 20-week treatment

After 20 weeks of riluzole treatment, correlation analysis of CEST signals and concentrations of Glu and GABA by 1H-MRS in the hippocampus and cortex of four mouse groups was performed (Fig. 5). GluCEST signals positively correlated with Glu concentrations in both hippocampus and cortex (R² = 0.709 and 0.660 respectively; p < 0.0001) (Fig. 5A), indicating consistency between CEST imaging and 1H-MRS in the detection of Glu levels. However, correlations between GABACEST signals and GABA concentrations in the hippocampus and cortex were not significant (R² = 0.011 and 0.03, p = 0.568 and 0.350, respectively).

Fig. 5
figure 5

Correlations analysis among CEST/MRS measurements of Glu/GABA and cognition after 20-week treatment. (A) Correlation analysis between GluCEST and GABACEST signals and their concentrations by 1H-MRS in the hippocampus and cortex for four mouse groups. (B) Correlation analysis between MWM behavior metrics and GluCEST/GABACEST signals in the hippocampus and cortex. (C) Correlation analysis between MWM behavior metrics and Glu concentrations by 1H-MRS in the hippocampus and cortex

Full size image

GluCEST signals strongly correlated with MWM performance, including target quadrant time ratios (R² = 0.683, p < 0.0001) and distance ratios (R² = 0.604, p < 0.0001) in the hippocampus, and similar correlations were observed in the cortex (time: R² = 0.514; distance: R² = 0.475, both p < 0.0001) (Fig. 5B). These findings indicated that higher Glu levels were associated with enhanced maze navigation performance in 3xTg AD mice. The correlation between GABACEST signals and MWM performance was weak or very weak. Glu levels detected by 1H-MRS showed a positive correlation with MWM performance (Fig. 5C), with stronger associations in the hippocampus (time: R² = 0.725; distance: R² = 0.577, both p < 0.0001) than in the cortex (time: R² = 0.354, p = 0.0004; distance: R² = 0.396, p = 0.0002). There was no significant correlation between GABA concentration measured by 1H-MRS and MWM performance (data not shown). These results highlighted that elevated Glu levels in the hippocampus of 3xTg AD mice were significantly correlated with the improvement of cognition and learning ability.

Effect of riluzole on pathological changes of brain in 3xTg AD mice

After 20 weeks of treatment with riluzole, immunohistochemistry and Nissl staining were used to evaluate the pathological changes of the brain in 3xTg AD mice. The areas of Aβ plaque in hippocampus and cortex in AD_C group were significantly larger than those in AD_R (hippocampus: p = 0.002; cortex: p = 0.001), C57_R and C57_C groups (p < 0.0001 for both regions) (Fig. 6A).

Fig. 6
figure 6

Effects of riluzole on pathological changes of brain in 3xTg AD mice. (A-D, left) Pathological images of brain sections in AD_R (n = 4), AD_C (n = 4), C57_R (n = 3), and C57_C (n = 3) groups after 20 weeks of riluzole treatment. (A-D, right) The corresponding quantitative analysis of Aβ staining, Tau staining, GFAP staining, and Nissl staining in the hippocampal and/or cortical regions. Data were presented as mean ± SD. Statistical significance was denoted by *** p < 0.001, **** p < 0.0001, # p < 0.05, ## p < 0.01, ### p < 0.001, and #### p < 0.0001, relative to C57 (C57_R and C 57_C) mice; $ p < 0.05 and $$ p < 0.01, relative to AD (AD_R or AD_C) mice

Full size image

Figure 6B showed that Tau protein immunoreactivity in hippocampus in AD_C group was significantly higher than that in AD_R group (p < 0.0001), but the latter was still significantly higher than that in C57_R group (p = 0.023) and C57_C group (p = 0.020) in healthy controls. In contrast, no significant positive tau staining was detected in the cortical regions at this age (Figure S3).

As shown in Fig. 6C, the hippocampus GFAP positive areas were significantly increased in AD_C group, compared to the AD_R (p = 0.0001), C57_R (p < 0.0001), and C57_C (p < 0.0001) groups. Similarly, the AD_C group had a significant increase in GFAP-positive areas of the cortex, compared to the other groups (all p < 0.0001).

Nissl staining results (Fig. 6D) showed that there were fewer neurons in hippocampus in AD_C group compared to other groups (vs. AD_R: p = 0.009; vs. C57_R: p = 0.0001; vs. C57_C: p = 0.0001). In the cortex, the number of neurons in the AD_C group was lower than in the C57_R and C57_C groups (both p = 0.0003).

Effects of riluzole on EAAT2, GAD 65/67, and GS protein expressions in 3xTg AD mice

After 20 weeks of treatment with riluzole, the expression levels of EAAT2, GAD 65/67 and GS in brain homogenates of 3xTg AD mice were analyzed by western blot (Fig. 7). As shown in Fig. 7A, levels of EAAT2 in the hippocampus were significantly lower in the AD_C group compared to all other groups (p < 0.0001). The levels of EAAT2 in hippocampus in AD_R group were significantly higher than those in AD_C group (p < 0.0001), but slightly lower than those in C57_R and C57_C groups (both p = 0.010). In the cortex, EAAT2 expression levels in the AD_C group were significantly lower than those in the other groups (p < 0.0001), but the levels in the AD_R group was comparable to that in the control group. Figure 7B shown that the expression of GAD 65/67 in hippocampus in AD_C group was lower than that of C57_R group (p = 0.001) and C57_C group (p = 0.002), and slightly higher in AD_R group than that of AD_C group (p = 0.035). In the cortex, the expression of GAD 65/67 in AD_C group was significantly lower than in C57_R group (p = 0.003) and C57_C group (p = 0.004), while the levels of AD_R group were similar to those of control groups. The expression of GS in hippocampus and cortex in AD_C group was significantly lower than that of C57_R group and C57_C group (both p < 0.0001), and there was no significant difference between AD_R group and healthy control group (Fig. 7C). These findings suggested that riluzole treatment partially restored the expression levels of EAAT2, GAD 65/67 and GS in AD mice, bringing them close to the levels in healthy controls.

Fig. 7
figure 7

Effects of riluzole on EAAT2, GAD 65/67, and GS protein expressions in 3xTg AD mice. (A-C, left) Western blot images of EAAT2, GAD 65/67, and GS expressions in the hippocampal and cortical regions of four mouse groups after 20 weeks of riluzole treatment. GAPDH was used as a loading control. (A-C, right) The corresponding quantitative analysis of each protein in the hippocampal and cortical regions of four mouse groups. Data were presented as mean ± SD. Statistical significance was indicated by *p < 0.05, **p < 0.01, ####p < 0.0001(relative to other groups), and $p < 0.05 (relative to the AD_C group)

Full size image

Discussion

In this study, CEST and 1H-MRS techniques were used for the first time to dynamically monitor the therapeutic effects of riluzole on the changes of Glu and GABA levels in the 3xTg AD mouse model. After 20 weeks of treatment, riluzole normalized the signal intensity of GluCEST and GABACEST, as well as Glu and GABA concentrations measured by 1H-MRS, to control mice lebels in 3xTg AD mice brains. Moreover, riluzole significantly improved cognitive function and alleviated AD-related pathological changes. Importantly, a strong correlation was established between GluCEST signals, Glu levels measured by 1H-MRS, and cognitive performance, emphasizing the potential of riluzole in improving neurotransmitter metabolic disorders in the brain of AD mice.

Untreated 3xTg AD mice had significantly lower levels of Glu and GABA quantified by 1H-MRS (Fig. 2) and reduced GluCEST and GABACEST signals in the hippocampus and cortex (Fig. 3) compared to controls at 20 weeks post-treatment. These findings aligned with previous reported studies of reduced Glu levels in AD patients, especially in regions associated with memory and cognition [33,34,35], suggested potential disruptions in neurotransmitter metabolisms in AD pathogenesis [6]. Furthermore, the strong correlation between GluCEST signals and Glu concentrations quantified by 1H-MRS (Fig. 5A), confirmed the reliability of CEST imaging to reflect the changes of AD-related Glu. Interestingly, GluCEST signals in the hippocampus were more strongly correlated with cognitive performance than in the cortex (Fig. 5b), possibly due to the fact that the hippocampus plays a more direct role in memory formation and consolidation [36, 37], whereas cortical regions integrate various cognitive processes. However, the correlation between GABACEST signals and GABA concentrations measured by 1H-MRS was not significant, which may be due to the lower brain concentration of GABA and the reduced sensitivity of the PRESS sequence used for GABA detection [38]. The weak correlation between GABACEST signals and behavioral test results (Fig. 5A) suggests that although GABA is essential for maintaining excitation/inhibition balance and modulating synaptic plasticity, GABA may be less directly associated with specific cognitive functions, such as learning and memory, than Glu [39, 40]. 1H-MRS studies have consistently revealed abnormal changes in their levels in AD brains, correlating with the severity of cognitive impairment [41]. Dramatically decreased GABA has been reported in posterior cingulate and parietal cortex of AD patients [8, 42]. Although some results have been controversial [41, 43], our CEST and 1H-MRS results of reduced Glu and GABA levels in 3xTg AD mice, which were significantly linked to decreased cognitive and learning abilities (Fig. 5B), indicated that neurotransmitter systems dysfunction might be a major mechanism underlying AD pathogenesis and cognitive deficit [8, 44].

The levels of Glu and GABA in the brain of AD mice were significantly increased after treatment with riluzole. Enhanced GluCEST signals in the hippocampus and cortex (Fig. 3) and elevated concentrations of Glu quantified by 1H-MRS in the corresponding brain regions (Fig. 2), suggested that the neuroprotective effects of riluzole may involve improved balance and homeostasis of Glu [12, 45]. While GABACEST signals showed a trend towards increased GABA levels in riluzole-treated AD mice, particularly in the early stages of treatment. These findings indicate that riluzole’s effects on GABA levels may be more modest or variable compared to its impact on Glu homeostasis. The increased Glu and GABA levels we observed in riluzole-treated 3xTg-AD mice may initially seem to conflict with previous studies reporting riluzole-mediated reductions in glutamate release and signaling [13, 14]. However, it is important to note that our CEST and 1H-MRS measurements reflect total tissue concentrations of Glu and GABA, which are influenced by both presynaptic release and postsynaptic neuronal uptake. In contrast, earlier studies focused specifically on riluzole’s ability to decrease presynaptic glutamate release. We propose that the elevated Glu and GABA signals detected in our study may be due to an enhanced balance between synaptic release and reuptake, leading to the restoration of neurotransmitter homeostasis, which aligned with evidence that riluzole reduces presynaptic glutamate release while increasing postsynaptic glutamate uptake and recycling [17]. Interestingly, while the longitudinal changes in cortical GABA levels were less pronounced than those in the hippocampus, the cross-sectional comparison at the 20-week timepoint revealed significantly lower cortical GABA levels in the AD_C group compared to the C57_R and C57_C control groups. This finding suggested that riluzole treatment may also have a protective effect on GABA levels in the cortex, preventing the decline observed in AD mice. Supporting this concept, our behavioral and pathological findings demonstrated that the riluzole-induced changes in hippocampal Glu and GABA were associated with improvements in cognitive function and reductions in AD-related neuropathology. Although there is limited literature on the mechanism of action of riluzole on Glu and GABA metabolism in AD, studies have reported that riluzole can reduce Glu-related neurotoxicity and oxidative stress by influencing the WNT/β-catenin signaling pathway [45]. Riluzole has also been reported to enhance inhibitory neuronal GABAergic function, which may contribute to its neuroprotective effects in AD models [17]. Our noninvasive CEST and 1H-MRS results provided evidence of elevated Glu and GABA levels in riluzole-treated AD mice, further deepening our understanding of its mechanisms of action.

Treatment with riluzole significantly improved spatial learning and memory in AD mice, as demonstrated by reduced latency in finding the hidden platform and increased swimming time and distance in the MWM target quadrant (Fig. 4), consistent with previous studies [18, 19]. While riluzole-treated AD mice showed a trend towards normalized cognitive performance, the differences between the AD_R and control groups did not reach statistical significance at the 20-week timepoint. Nonetheless, the observed improvements in the AD_R group compared to the AD_C group suggested that riluzole may have potential therapeutic effects on cognitive function in AD. Importantly, CEST and 1H-MRS showed that cognitive performance after riluzole treatment was significantly associated with elevated Glu levels (Fig. 5), suggesting that restoration of the glutamatergic system might be a crucial mechanism through which riluzole ameliorates cognitive impairments in AD [13, 14]. Although limited by the relatively small sample size, the positive correlations observed between GluCEST signals, Glu concentrations, and cognitive performance are consistent with previous reports [12, 23]. These findings provide preliminary support for the potential of CEST imaging and 1H-MRS as tools for monitoring AD-related cognitive decline and evaluating therapeutic interventions.

Riluzole treatment reduced Aβ plaque deposition, alleviated tau hyperphosphorylation, weakened astrocyte activation, and mitigated neuroinflammation in the hippocampus and cortex (Fig. 6), further supporting the neuroprotective effects of riluzole observed in vivo through CEST imaging and 1H-MRS in 3xTg AD mice. Moreover, riluzole-treated AD mice exhibited higher neuronal counts compared to untreated controls, indicating its ability to prevent neuron loss, consistent with the previously reported [13, 19, 46]. The mechanism of neuroprotective effect of riluzole may involve its ability to modulate Glu excitotoxicity and restore Glu/GABA homeostasis, as demonstrated by the increased Glu and GABA levels detected by CEST and 1H-MRS. Excessive glutamatergic neurotransmission has been found to upregulate the activity of the major Aβ-producing BACE1, and activate kinases such as glycogen synthase kinase 3β(GSK3β), which are involved in the tau hyperphosphorylation [47, 48]. By restoring neurotransmitter balance, riluzole may indirectly inhibit these pathological processes, thereby reducing Aβ plaque deposition and tau hyperphosphorylation. In addition, riluzole’s ability to maintain Glu/GABA homeostasis and inhibit Glu excitotoxicity may also contribute to its anti-inflammatory effects by attenuating excessive Glu-induced inflammatory responses and glial cell activation [7].

Riluzole increased the expression levels of EAAT2, GAD65/67 and GS, the key proteins involved in Glu and GABA metabolism, in the brain of 3xTg AD mice (Fig. 7). Upregulation of EAAT2, the major Glu transporter, suggests enhanced Glu uptake and homeostasis, contributing to neuroprotection [49,50,51]. Increased expressions of GAD65/67 indicated improved GABA synthesis and restoration of the excitatory/inhibitory balance [17, 52,53,54]. Furthermore, while the difference in GS expression between the AD_R and AD_C groups did not reach statistical significance, GS levels were significantly reduced in the AD_C group compared to the C57_R and C57_C groups. The trend towards increased GS expression in the AD_R group suggests that riluzole may have a positive effect on GS levels, potentially contributing to enhanced conversion of Glu to glutamine and reducing the risk of glutamate accumulation and excitotoxicity [54, 55]. These molecular changes provided a mechanistic basis for the observed alterations in neurotransmitter levels detected by CEST and 1H-MRS.

Limitations

This study has some limitations that should be considered. First, a single pathological timepoint restricts the observation of dynamic changes, and the effects of a single drug on the complex pathogenesis of AD may be limited. Second, despite the advantages of CEST and 1H-MRS in detecting neurotransmitter changes, they present limitations in the accuracy and specificity of quantitative analysis due to the proximity of the chemical exchange frequencies of Glu and GABA [32]. It is also important to note the potential spectral overlap between GABA and glutamate in CEST imaging, which may impact the specificity of GABA measurements at 2.75 ppm [30, 32]. Future studies could employ advanced acquisition and analysis techniques to further mitigate this issue and improve the specificity of GABA measurements [56, 57]. Additionally, the PRESS sequence used for 1H-MRS in this study has inherent limitations for GABA quantification. The relatively low concentration of GABA in the brain, combined with potential spectral overlap with other metabolites, can lead to uncertainty in absolute GABA quantification. Specialized sequences optimized for GABA detection, such as MEGA-PRESS, could provide more reliable and specific measurements [58].They provided overall measurement rather than specific insights into neurotransmitter function [59]. Furthermore, the relatively small sample size in each group may limit the statistical power and reliability of the reported associations. This necessitates cautious interpretation of the findings and highlights the need for validation studies in larger cohorts. However, the consistency of our findings with published studies reinforces the reliability and reproducibility of these technologies in reflecting AD-related metabolic changes [25, 44].

Conclusion

CEST imaging combined with 1H-MRS enables non-invasive, dynamic, and quantitative monitoring of neurotransmitter alterations in AD brain. The strong correlation between MR-based metabolic findings, behavioral and pathological results further support the potential of CEST and 1H-MRS as biomarkers for the assessment of the efficacy of personalized AD therapy. This study reveals the neuroprotective mechanism of riluzole in restoring Glu and GABA homeostasis, reducing pathological changes, and improving cognitive function in 3xTg AD mice. As an FDA-approved drug with a well-established safety profile, riluzole has significant potential for clinical translation in AD management. Future studies should focus on optimizing these imaging techniques and translating these findings to clinical applications.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

1H-MRS:

Proton magnetic resonance spectroscopy

Aβ:

Amyloid-beta

AD:

Alzheimer’s disease

ALS:

Amyotrophic lateral sclerosis

ANOVA:

One-way analysis of variance

ARRIVE:

Animal Research: Reporting of In Vivo Experiments

BCA:

Bicinchoninic acid

CEST:

Chemical exchange saturation transfer

Cr:

Creatine

EAAT2:

Excitatory amino acid transporter 2

ECL:

Enhanced chemiluminescence

FDA:

Food and Drug Administration

GABA:

Gamma-aminobutyric acid

GABACEST:

GABA-weighted-CEST

GAD65/67:

Glutamic acid decarboxylase 65/67

GAPDH:

Glyceraldehyde 3-phosphate dehydrogenase

GFAP:

Glial fibrillary acidic protein

Gln:

Glutamine

Glu:

Glutamate

GluCEST:

Glu-weighted-CEST

GRE:

Gradient recalled echo

GS:

Glutamine synthetase

HPF:

High-power field of view

HRP:

Horseradish peroxidase

Ins:

Myo-inositol

MWM:

Morris Water Maze

NAA:

N-acetylaspartate

PRESS:

Point resolved spectroscopy sequence

PVDF:

Polyvinylidene difluoride

RIPA:

Radioimmunoprecipitation assay

ROIs:

Regions of interest

Tau:

Taurine

Tau pS404:

Tau phosphorylated at serine 404

TBST:

Tris-buffered saline with 0.1% Tween 20

tCho:

Total choline

TE:

Echo time

TR:

Repetition time

VAPOR:

Variable angle spin echo with optimized relaxation delays

VOIs:

Volumes of interest

References

  1. 2024 Alzheimer’s disease facts and figures. Alzheimers Dement. 2024;20(5):3708 – 821.

  2. Nichols E, Vos TJAs D. The estimation of the global prevalence of dementia from 1990-2019 and forecasted prevalence through 2050: an analysis for the global burden of Disease (GBD) study 2019. 2021;17:e051496.

  3. Tahami Monfared AA, Byrnes MJ, White LA, Zhang Q. Alzheimer’s Disease: Epidemiology and Clinical Progression. Neurol Ther. 2022;11(2):553–69.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Estimation of the global prevalence of dementia. In 2019 and forecasted prevalence in 2050: an analysis for the global burden of Disease Study 2019. Lancet Public Health. 2022;7(2):e105–25.

    Article  Google Scholar 

  5. Nandi A, Counts N, Chen S, Seligman B, Tortorice D, Vigo D, et al. Global and regional projections of the economic burden of Alzheimer’s disease and related dementias from 2019 to 2050: a value of statistical life approach. EClinicalMedicine. 2022;51:101580.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Long JM, Holtzman DM. Alzheimer Disease: an update on pathobiology and treatment strategies. Cell. 2019;179(2):312–39.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  7. Wu M, Zhang M, Yin X, Chen K, Hu Z, Zhou Q, et al. The role of pathological tau in synaptic dysfunction in Alzheimer’s diseases. Translational Neurodegeneration. 2021;10(1):45.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  8. Carello-Collar G, Bellaver B, Ferreira PCL, Ferrari-Souza JP, Ramos VG, Therriault J, et al. The GABAergic system in Alzheimer’s disease: a systematic review with meta-analysis. Mol Psychiatry. 2023;28(12):5025–36.

    Article  PubMed  CAS  Google Scholar 

  9. Passeri E, Elkhoury K, Morsink M, Broersen K, Linder M, Tamayol A et al. Alzheimer’s Disease: Treatment Strategies and Their Limitations. Int J Mol Sci. 2022;23(22).

  10. Reardon S. Alzheimer’s drug with modest benefits wins backing of FDA advisers. Nature. 2024.

  11. Yu TW, Lane HY, Lin CH. Novel Therapeutic Approaches for Alzheimer’s Disease: An Updated Review. Int J Mol Sci. 2021;22(15).

  12. Matthews DC, Mao X, Dowd K, Tsakanikas D, Jiang CS, Meuser C, et al. Riluzole, a glutamate modulator, slows cerebral glucose metabolism decline in patients with Alzheimer’s disease. Brain. 2021;144(12):3742–55.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Hunsberger HC, Weitzner DS, Rudy CC, Hickman JE, Libell EM, Speer RR, et al. Riluzole rescues glutamate alterations, cognitive deficits, and tau pathology associated with P301L tau expression. J Neurochem. 2015;135(2):381–94.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  14. Hascup KN, Findley CA, Britz J, Esperant-Hilaire N, Broderick SO, Delfino K, et al. Riluzole attenuates glutamatergic tone and cognitive decline in AβPP/PS1 mice. J Neurochem. 2021;156(4):513–23.

    Article  PubMed  CAS  Google Scholar 

  15. Pfitzer J, Pinky PD, Perman S, Redmon E, Cmelak L, Suppiramaniam V et al. Troriluzole rescues glutamatergic deficits, amyloid and tau pathology, and synaptic and memory impairments in 3xTg-AD mice. J Neurochem. 2024.

  16. Saba K, Patel AB. Riluzole restores memory and brain energy metabolism in AβPP-PS1 mouse model of Alzheimer’s disease. Biochem Biophys Res Commun. 2022;610:140–6.

    Article  PubMed  CAS  Google Scholar 

  17. Yang Y, Ji WG, Zhang YJ, Zhou LP, Chen H, Yang N, et al. Riluzole ameliorates soluble Aβ(1–42)-induced impairments in spatial memory by modulating the glutamatergic/GABAergic balance in the dentate gyrus. Prog Neuro-psychopharmacol Biol Psychiatry. 2021;108:110077.

    Article  CAS  Google Scholar 

  18. Mokhtari Z, Baluchnejadmojarad T, Nikbakht F, Mansouri M, Roghani M. Riluzole ameliorates learning and memory deficits in Aβ25-35-induced rat model of Alzheimer’s disease and is independent of cholinoceptor activation. Biomed Pharmacotherapy = Biomedecine Pharmacotherapie. 2017;87:135–44.

    Article  PubMed  CAS  Google Scholar 

  19. Okamoto M, Gray JD, Larson CS, Kazim SF, Soya H, McEwen BS, et al. Riluzole reduces amyloid beta pathology, improves memory, and restores gene expression changes in a transgenic mouse model of early-onset Alzheimer’s disease. Translational Psychiatry. 2018;8(1):153.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Song T, Song X, Zhu C, Patrick R, Skurla M, Santangelo I, et al. Mitochondrial dysfunction, oxidative stress, neuroinflammation, and metabolic alterations in the progression of Alzheimer’s disease: a meta-analysis of in vivo magnetic resonance spectroscopy studies. Ageing Res Rev. 2021;72:101503.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Huang D, Liu D, Yin J, Qian T, Shrestha S, Ni H. Glutamate-glutamine and GABA in brain of normal aged and patients with cognitive impairment. Eur Radiol. 2017;27(7):2698–705.

    Article  PubMed  Google Scholar 

  22. Londono AC, Castellanos FX, Arbelaez A, Ruiz A, Aguirre-Acevedo DC, Richardson AM, et al. An 1H-MRS framework predicts the onset of Alzheimer’s disease symptoms in PSEN1 mutation carriers. Alzheimers Dement. 2014;10(5):552–61.

    Article  PubMed  Google Scholar 

  23. Wong D, Atiya S, Fogarty J, Montero-Odasso M, Pasternak SH, Brymer C, et al. Reduced hippocampal glutamate and posterior cingulate N-Acetyl aspartate in mild cognitive impairment and Alzheimer’s disease is Associated with episodic memory performance and White Matter Integrity in the Cingulum: a pilot study. J Alzheimer’s Disease: JAD. 2020;73(4):1385–405.

    Article  PubMed  Google Scholar 

  24. Crescenzi R, DeBrosse C, Nanga RP, Byrne MD, Krishnamoorthy G, D’Aquilla K, et al. Longitudinal imaging reveals subhippocampal dynamics in glutamate levels associated with histopathologic events in a mouse model of tauopathy and healthy mice. Hippocampus. 2017;27(3):285–302.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  25. Crescenzi R, DeBrosse C, Nanga RP, Reddy S, Haris M, Hariharan H, et al. In vivo measurement of glutamate loss is associated with synapse loss in a mouse model of tauopathy. NeuroImage. 2014;101:185–92.

    Article  PubMed  CAS  Google Scholar 

  26. Igarashi H, Ueki S, Kitaura H, Kera T, Ohno K, Ohkubo M et al. Longitudinal GluCEST MRI Changes and Cerebral Blood Flow in 5xFAD Mice. Contrast media & molecular imaging. 2020;2020:8831936.

  27. Lee DH, Lee DW, Kwon JI, Kim ST, Woo CW, Kon Kim J, et al. Changes to gamma-aminobutyric acid levels during short-term epileptiform activity in a kainic acid-induced rat model of status epilepticus: a chemical exchange saturation transfer imaging study. Brain Res. 2019;1717:176–81.

    Article  PubMed  CAS  Google Scholar 

  28. Liu Y, Zong X, Huang J, Guan Y, Li Y, Du T, et al. Ginsenoside Rb1 regulates prefrontal cortical GABAergic transmission in MPTP-treated mice. Aging. 2019;11(14):5008–34.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  29. Orzyłowska A, Oakden W. Saturation Transfer MRI for Detection of Metabolic and Microstructural Impairments Underlying Neurodegeneration in Alzheimer’s Disease. Brain Sci. 2021;12(1).

  30. Yan G, Zhang T, Dai Z, Yi M, Jia Y, Nie T, et al. A potential magnetic resonance imaging technique based on Chemical Exchange Saturation transfer for in vivo γ-Aminobutyric acid imaging. PLoS ONE. 2016;11(10):e0163765.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Kim M, Gillen J, Landman BA, Zhou J, van Zijl PC. Water saturation shift referencing (WASSR) for chemical exchange saturation transfer (CEST) experiments. Magn Reson Med. 2009;61(6):1441–50.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Cai K, Haris M, Singh A, Kogan F, Greenberg JH, Hariharan H, et al. Magnetic resonance imaging of glutamate. Nat Med. 2012;18(2):302–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  33. Haris M, Nath K, Cai K, Singh A, Crescenzi R, Kogan F, et al. Imaging of glutamate neurotransmitter alterations in Alzheimer’s disease. NMR Biomed. 2013;26(4):386–91.

    Article  PubMed  CAS  Google Scholar 

  34. Oeltzschner G, Wijtenburg SA, Mikkelsen M, Edden RAE, Barker PB, Joo JH, et al. Neurometabolites and associations with cognitive deficits in mild cognitive impairment: a magnetic resonance spectroscopy study at 7 Tesla. Neurobiol Aging. 2019;73:211–8.

    Article  PubMed  CAS  Google Scholar 

  35. Rupsingh R, Borrie M, Smith M, Wells JL, Bartha R. Reduced hippocampal glutamate in Alzheimer disease. Neurobiol Aging. 2011;32(5):802–10.

    Article  PubMed  CAS  Google Scholar 

  36. Bellanti F, Bukke VN, Moola A, Villani R, Scuderi C, Steardo L, et al. Effects of Ultramicronized Palmitoylethanolamide on mitochondrial bioenergetics, cerebral metabolism, and glutamatergic transmission: an Integrated Approach in a Triple Transgenic Mouse Model of Alzheimer’s Disease. Front Aging Neurosci. 2022;14:890855.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  37. Yeung JHY, Walby JL, Palpagama TH, Turner C, Waldvogel HJ, Faull RLM, et al. Glutamatergic receptor expression changes in the Alzheimer’s disease hippocampus and entorhinal cortex. Brain Pathol. 2021;31(6):e13005.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  38. Mullins PG, McGonigle DJ, O’Gorman RL, Puts NA, Vidyasagar R, Evans CJ, et al. Current practice in the use of MEGA-PRESS spectroscopy for the detection of GABA. NeuroImage. 2014;86:43–52.

    Article  PubMed  CAS  Google Scholar 

  39. Andersen JV, Schousboe A, Verkhratsky A. Astrocyte energy and neurotransmitter metabolism in Alzheimer’s disease: integration of the glutamate/GABA-glutamine cycle. Prog Neurobiol. 2022;217:102331.

    Article  PubMed  CAS  Google Scholar 

  40. Hone-Blanchet A, Bohsali A, Krishnamurthy LC, Shahid SS, Lin Q, Zhao L, et al. Frontal metabolites and Alzheimer’s disease biomarkers in healthy older women and women diagnosed with mild cognitive impairment. J Alzheimer’s Disease: JAD. 2022;87(3):1131–41.

    Article  PubMed  CAS  Google Scholar 

  41. Riese F, Gietl A, Zölch N, Henning A, O’Gorman R, Kälin AM, et al. Posterior cingulate γ-aminobutyric acid and glutamate/glutamine are reduced in amnestic mild cognitive impairment and are unrelated to amyloid deposition and apolipoprotein E genotype. Neurobiol Aging. 2015;36(1):53–9.

    Article  PubMed  CAS  Google Scholar 

  42. Bai X, Edden RA, Gao F, Wang G, Wu L, Zhao B, et al. Decreased γ-aminobutyric acid levels in the parietal region of patients with Alzheimer’s disease. J Magn Reson Imaging: JMRI. 2015;41(5):1326–31.

    Article  PubMed  Google Scholar 

  43. Andersen JV, Christensen SK, Westi EW, Diaz-delCastillo M, Tanila H, Schousboe A, et al. Deficient astrocyte metabolism impairs glutamine synthesis and neurotransmitter homeostasis in a mouse model of Alzheimer’s disease. Neurobiol Dis. 2021;148:105198.

    Article  PubMed  CAS  Google Scholar 

  44. Burnyasheva AO, Stefanova NA, Kolosova NG, Telegina DV. Changes in the Glutamate/GABA system in the Hippocampus of rats with age and during Alzheimer’s Disease signs Development. Biochem Biokhimiia. 2023;88(12):1972–86.

    Article  CAS  Google Scholar 

  45. Vallée A, Vallée JN, Guillevin R, Lecarpentier Y. Riluzole: a therapeutic strategy in Alzheimer’s disease by targeting the WNT/β-catenin pathway. Aging. 2020;12(3):3095–113.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Pereira AC, Gray JD, Kogan JF, Davidson RL, Rubin TG, Okamoto M, et al. Age and Alzheimer’s disease gene expression profiles reversed by the glutamate modulator riluzole. Mol Psychiatry. 2017;22(2):296–305.

    Article  PubMed  CAS  Google Scholar 

  47. Tanaka T, Ohashi S, Takashima A, Kobayashi S. Glutamate-responsive translation of dendritic GSK3β mRNA triggers a cycle for amplification of reactivated preexisting GSK3β that is indispensable for tau hyperphosphorylation. Neurochem Int. 2020;139:104808.

    Article  PubMed  Google Scholar 

  48. Zott B, Simon MM, Hong W, Unger F, Chen-Engerer HJ, Frosch MP, et al. A vicious cycle of β amyloid-dependent neuronal hyperactivation. Sci (New York NY). 2019;365(6453):559–65.

    Article  CAS  Google Scholar 

  49. Banasr M, Chowdhury GM, Terwilliger R, Newton SS, Duman RS, Behar KL, et al. Glial pathology in an animal model of depression: reversal of stress-induced cellular, metabolic and behavioral deficits by the glutamate-modulating drug riluzole. Mol Psychiatry. 2010;15(5):501–11.

    Article  PubMed  CAS  Google Scholar 

  50. Kim K, Lee SG, Kegelman TP, Su ZZ, Das SK, Dash R, et al. Role of excitatory amino acid transporter-2 (EAAT2) and glutamate in neurodegeneration: opportunities for developing novel therapeutics. J Cell Physiol. 2011;226(10):2484–93.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  51. Masliah E, Alford M, DeTeresa R, Mallory M, Hansen L. Deficient glutamate transport is associated with neurodegeneration in Alzheimer’s disease. Ann Neurol. 1996;40(5):759–66.

    Article  PubMed  CAS  Google Scholar 

  52. Buddhala C, Hsu CC, Wu JY. A novel mechanism for GABA synthesis and packaging into synaptic vesicles. Neurochem Int. 2009;55(1–3):9–12.

    Article  PubMed  CAS  Google Scholar 

  53. Schwab C, Yu S, Wong W, McGeer EG, McGeer PL. GAD65, GAD67, and GABAT immunostaining in human brain and apparent GAD65 loss in Alzheimer’s disease. J Alzheimer’s Disease: JAD. 2013;33(4):1073–88.

    Article  PubMed  CAS  Google Scholar 

  54. Olabarria M, Noristani HN, Verkhratsky A, Rodríguez JJ. Age-dependent decrease in glutamine synthetase expression in the hippocampal astroglia of the triple transgenic Alzheimer’s disease mouse model: mechanism for deficient glutamatergic transmission? Mol Neurodegeneration. 2011;6:55.

    Article  CAS  Google Scholar 

  55. Rose CF, Verkhratsky A, Parpura V. Astrocyte glutamine synthetase: pivotal in health and disease. Biochem Soc Trans. 2013;41(6):1518–24.

    Article  PubMed  CAS  Google Scholar 

  56. Maguin C, Mougel E, Valette J, Flament J. Toward quantitative CEST imaging of glutamate in the mouse brain using a multi-pool exchange model calibrated by (1)H-MRS. Magnetic resonance in medicine. 2024.

  57. Severo F. Shemesh NJapa. In-vivo Magnetic Resonance Imaging of GABA and Glutamate. 2020.

  58. Mikkelsen M, Barker PB, Bhattacharyya PK, Brix MK, Buur PF, Cecil KM, et al. Big GABA: edited MR spectroscopy at 24 research sites. NeuroImage. 2017;159:32–45.

    Article  PubMed  CAS  Google Scholar 

  59. Rae CD. A guide to the metabolic pathways and function of metabolites observed in human brain 1H magnetic resonance spectra. Neurochem Res. 2014;39(1):1–36.

    Article  PubMed  CAS  Google Scholar 

Download references

Funding

This study was supported by grants from the National Natural Science Foundation of China (82071973 to Y.L., 82020108016 to R.H.W.), Guangdong Basic and Applied Basic Research Foundation (2023A1515010326 to Y.L., 2020A1515011022 to Y.L.), Key Research Platform and Project of Guangdong University (2022ZDZX2020 to Y.L.), Shantou Science and Technology Project (240428226498013 to Y.Y.S.) and Swiss National Science Foundation (No. 213769 to L.J.X.).

Author information

Authors and Affiliations

  1. Radiology Department, Second Affiliated Hospital of Shantou University Medical College, Shantou, Guangdong, China

    Yuanyu Shen, Xiaolei Zhang, Siqi Liu, Wentao Xuan, Caiyu Zhuang, Yue Chen, Beibei Chen, Xinhui Zheng, Renhua Wu & Yan Lin

  2. Center for Biomedical Imaging (CIBM), Lausanne, Switzerland

    Lijing Xin

  3. Animal Imaging and Technology, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland

    Lijing Xin

  4. Institute of Physics (IPHYS), Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland

    Lijing Xin

Authors
  1. Yuanyu Shen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Xiaolei Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Siqi Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Lijing Xin
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Wentao Xuan
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Caiyu Zhuang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Yue Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Beibei Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Xinhui Zheng
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Renhua Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Yan Lin
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Y.L supervised and directed this project. Y.S., X.Z., S.L., C.Z., B.C, Y.L, and R.W. contributed to the conception and design of the study. Y.S., S.L., L.X., X.Z, W.X., Y.C. and X.Z. contributed to the acquisition and analysis of data. Y.S. contributed to drafting the text and providing the figures. Y.L contributed to the interpretation of the results and critical revision of the manuscript for important intellectual content and approved the final version of the manuscript. All the authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Yan Lin.

Ethics declarations

Ethics approval and consent to participate

All animal experiments were approved by the Ethics Committee of Shantou University Medical College (Approval ID: SUMC2022-597) and conducted in accordance with ARRIVE guidelines.

Consent of publication

All authors have read and agreed with the publication of the manuscript in its current form.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shen, Y., Zhang, X., Liu, S. et al. CEST imaging combined with 1H-MRS reveal the neuroprotective effects of riluzole by improving neurotransmitter imbalances in Alzheimer’s disease mice. Alz Res Therapy 17, 20 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-025-01672-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-025-01672-3

Keywords

Alzheimer's Research & Therapy

ISSN: 1758-9193