Increased glutamate and glutamine levels and their relationship to astrocytes and dopaminergic transmissions in the brains of adults with autism

Scientific Reports volume 13, Article number: 11655 (2023) Cite this article

785 Accesses

1 Citations

8 Altmetric

Metrics details

Increased excitatory neuronal tones have been implicated in autism, but its mechanism remains elusive. The amplified glutamate signals may arise from enhanced glutamatergic circuits, which can be affected by astrocyte activation and suppressive signaling of dopamine neurotransmission. We tested this hypothesis using magnetic resonance spectroscopy and positron emission tomography scan with 11C-SCH23390 for dopamine D1 receptors in the anterior cingulate cortex (ACC). We enrolled 18 male adults with high-functioning autism and 20 typically developed (TD) male subjects. The autism group showed elevated glutamate, glutamine, and myo-inositol (mI) levels compared with the TD group (p = 0.045, p = 0.044, p = 0.030, respectively) and a positive correlation between glutamine and mI levels in the ACC (r = 0.54, p = 0.020). In autism and TD groups, ACC D1 receptor radioligand binding was negatively correlated with ACC glutamine levels (r = − 0.55, p = 0.022; r = − 0.58, p = 0.008, respectively). The enhanced glutamate-glutamine metabolism might be due to astroglial activation and the consequent reinforcement of glutamine synthesis in autistic brains. Glutamine synthesis could underly the physiological inhibitory control of dopaminergic D1 receptor signals. Our findings suggest a high neuron excitation-inhibition ratio with astrocytic activation in the etiology of autism.

Autism is a developmental condition characterized by deficits in social communication and interaction, restricted interest, and repetitive behaviors. Autism has been associated with a high rate of suicide1 and impaired quality of life2. A survey in 2018 revealed that the autism spectrum disorder prevalence was 2.3% higher than previously reported3. Despite these critical issues, drug development to treat autism remains challenging4; understanding the pathophysiology of autism at the cellular and neurochemical levels is needed for finding novel targets to develop therapeutic treatments.

In the brain, assemblies of excitatory and inhibitory neurons are involved in learning and memory and the balance between their activities is critically significant for their network development5. One of the plausible theories for the pathophysiology of autism is the excitation-inhibition (E/I) imbalance theory6. Accordingly, increased E/I ratio has been implicated in autism7,8. At a non-clinical level, a mouse model of autism was reported to exhibit an elevated excitatory synaptic input in comparison to the inhibitory synaptic inputs9,10. In addition, optogenetic enhancement of the E/I ratio in the mouse neocortex provoked deficits in social behaviors relevant to autism11. In humans, disruption of inhibitory circuits putatively raised the E/I ratio as shown in electroencephalography12.

Magnetic resonance spectroscopy (MRS) allows to investigate the balance between the excitatory and inhibitory neuronal activities in the brains of living subjects, as this technology allows measuring excitatory glutamate (Glu), its relative metabolite glutamine (Gln), and the inhibitory gamma-aminobutyric acid (GABA) concentrations13,14,15. Only three studies which attempted to assay Glu and Gln levels separately by the conventional point resolved spectroscopy (PRESS) protocol in subjects with autism16,17,18. Among them, the two studies showed alterations in Glu18 and Gln16 levels in the anterior cingulate cortex (ACC). In contrast to conventional MRS protocols16,17,18, advanced methods offer more precise and sensitive detection of brain metabolites19, but they have so far not been applied to autism. Recently, we have demonstrated that a short-echo time (TE) spin-echo full-intensity acquisition localized single voxel spectroscopy (SPECIAL) sequence could robustly assess Gln levels in the brain, which correlated with plasma levels20. Accordingly, this approach might facilitate investigating the existence of an altered E/I ratio in the ACC of subjects with autism.

Although the mechanism underlying the augmented excitatory neuronal tone in autism is yet to be clarified, several lines of evidence support that astrocytes are involved in the modulation of the neuronal excitability in central nervous system diseases21,22, as astrocytes are involved in the regulation of Glu, Gln, and GABA levels in the brain23. For instance, Glu is converted to Gln by glutamine synthetase (GS), an enzyme found in astrocytes in the mammalian brain24. In fact, some studies have highlighted the contribution of astrocytes to the pathophysiology of autism25. Alternatively, human studies reported changes in the expression of astrocytic markers in postmortem brains as exemplified by increased glial fibrillary acidic protein (GFAP) and connexin 43 and decreased aquaporin 4 levels26. As such, elevated GFAP levels were detected in plasma samples derived from autistic patients27. Moreover, astrocytic modification of excitatory neurotransmitters is mechanistically linked to the dopaminergic system. Mice lacking dopamine (DA) D1 receptors (Rs) displayed increased Gln levels in the brain, an effect ameliorated by L-DOPA administration28. Moreover, autistic behaviors are hypothesized to arise from dopaminergic dysfunction29. Our recent study using positron emission tomography (PET) revealed the association between DA D1R radioligand binding in several brain regions with characteristic symptoms of adults with autism30. Therefore, DA signals mediated by D1Rs might be involved in aberrant Glu-Gln cycles and the consequent dysregulation of the neuronal E/I balance in autism. Despite these indications, no studies using in vivo imaging of human brain tissues have examined astrocyte activation, changes in the E/I balance in autism and its relationship with the dopaminergic nervous system.

Herein, we aimed at examining the hypothesis stating that the excitatory glutamatergic neurotransmissions were enhanced in the autistic brains in association with the astrocytic activation which is related to Gln synthesis and disrupted dopaminergic signals through D1Rs. Toward this end, we quantified Glu, Gln, and GABA levels by MRS to evaluate the excitatory versus the inhibitory neuronal tones, along with MRS assays of myo-inositol (mI), a marker for astrocytic activation. We also investigated the inter-metabolic correlations between mI and Gln levels. Therefore, we measured the metabolite levels and the regional DA D1R bindings with a specific PET radioligand of 11C-SCH23390. We focused on the ACC for its disrupted structural31,32 and functional33,34 properties as evidenced in the autistic brain. Participants were limited to adults with high-functioning autism, and males considering a potential gender difference in the E/I ratio35.

As previously mentioned, participants’ demographic and clinical characteristics are summarized in Table 1.

Glu levels were significantly increased in the ACC in adults with autism compared to TD subjects (12.93 [1.27] for the autism group, 12.07 [1.26] for the TD group, t = 2.08, p = 0.045), Gln (2.97 [0.39] for the autism group, 2.70 [0.41] for the TD group, t = 2.09, p = 0.044), and mI (8.42 [1.03] for the autism group, 7.57 [1.27] for the TD group; t = 2.26, p = 0.030) (Fig. 1). Alternatively, we found no significant differences in ACC GABA levels (2.38 [0.31] for the autism group, 2.29 [0.40] for the TD group; t = 0.79, p = 0.44).

Representative MRS spectrum and VOIs and scatterplots of metabolite levels in the ACC of subjects with autism and TD. (a) ACC VOI (30 × 20 × 20 mm3) of a TD individual. (b) Example of a MRS spectrum. (c) Glu, (d) Gln, and (e) mI levels in the ACC were significantly increased in autism compared with TD (Glu, p = 0.045; Gln, p = 0.044; mI, p = 0.030). ACC, anterior cingulate cortex; GABA, γ-aminobutyric acid; Gln, glutamine; Glu, glutamate; mI, myo-inositol; TD, typically developed; VOI, volume of interest. *p < 0.05.

We found significantly positive correlations between mI and Gln levels in the ACC of individuals with autism (r = 0.54, p = 0.020) but not in TD subjects (r = 0.091, p = 0.70) (Fig. 2). In addition, we found significantly positive correlations between Autism Spectrum Quotient (AQ)36, Attention switching subscale score, and Gln (r = 0.48, p = 0.045) and mI levels (r = 0.51, p = 0.029) in the ACC in individuals with autism (Fig. 3, Table S1).

mI and Gln levels in the ACC of subjects with autism and TD. (a) mI and Gln levels in the ACC were significantly correlated (r = 0.54, p = 0.020) in individuals with autism. (b) Conversely, no correlation was found in individuals with TD (r = 0.091, p = 0.70). ACC, anterior cingulate cortex; Gln, glutamine; mI, myo-inositol; TD, typical developed. *p < 0.05.

AQ Attention switching subscale score with Gln and mI levels in the ACC of individuals with autism. The AQ Attention switching subscale score was significantly positive correlated with (a) Gln levels (r = 0.48, p = 0.045) and with (b) mI levels (r = 0.51, p = 0.029) in the ACC in individuals with autism. ACC, anterior cingulate cortex; AQ, autism spectrum quotient; Gln, glutamine; mI, myo-inositol. *p < 0.05.

As reported previously30, there were no significant differences in radioligand binding to DA D1Rs between groups (0.36 [0.06] for the autism group, 0.38 [0.07] for the TD group). On the other hand, there was a negative correlation between DA D1R binding and Gln levels in the ACC (r = − 0.55, p = 0.022) of subjects with autism (Fig. 4). We also found negative correlations of Gln levels with DA D1R binding in the ACC (r = − 0.58, p = 0.008) of TD subjects (Fig. 4). In contrast, there were no significant correlations of DA D1R binding in any of these regions with mI and Glu levels in the ACC of any participants (p > 0.05) (Table S2).

Correlations of DA D1R binding with Gln levels in the ACC in individuals with autism and TD. Significant negative correlations between DA D1R binding and Gln levels in the ACC of individuals with (a) autism (r = − 0.55, p = 0.022) and (b) TD (r = − 0.58, p = 0.008). ACC, anterior cingulate cortex; DA D1R, dopamine D1 receptor; Gln, glutamine; TD, typical developed. *p < 0.05.

In this study, we found elevated Glu, Gln, and mI levels in the ACC of adults with autism compared with TD subjects. Subsequently, we found a positive correlation between Gln and mI levels in the ACC of subjects with autism but no correlation in TD subjects. 11C-SCH23390 PET also indicated that DA D1R binding was negatively correlated with Gln levels in the ACC in both adults with autism and TD subjects, but not with mI levels. With respect to clinical symptoms, subjects with autism showed a correlation between AQ Attention switching subscale scores and Gln and mI levels in the ACC.

The present study revealed that adults with autism exhibit increased excitatory Glu and Gln levels in the ACC. Our findings are supported by the evidence of neural hyperexcitability in autism7,8. While numerous MRS studies have attempted to clarify the E/I balance in autism15, only three separately evaluated Glu and Gln levels16,17,18. One reported increased Gln levels in the ACC16 while the others showed decreased Glu levels in the ACC18 and the striatum17. However, the SPECIAL sequence allowed us to evaluate Glu, Gln, and GABA levels separately and concurrently37. This advanced MRS protocol provided evidence of elevated levels of both excitatory neurotransmitters Glu and Gln in the ACC. As another feature of our study, the limited samples of male individuals with autism might contribute to revealing increased excitatory neurotransmitters, given the sex difference in the excitatory-inhibitory ratio in individuals with autism35. The relationship between autism and E/I imbalance has also received attention from a mechanistic viewpoint in genetic disorders. In fact, Fragile X syndrome, which is caused by changes in a single gene and is associated with autistic manifestations38,39, has shown aberrant E/I ratios according to several lines of reports (reviewed in40) in agreement with the present findings.

Consistent with previous studies41,42, adults with autism showed an increase in levels of mI43. Based on accumulated evidence from animal44,45 and human studies26,27 concerning the role of astrocytes in autism, altered astroglial functions have been hypothesized in order to contribute to the pathogenesis of autism25. Under metabolic homeostasis, astrocytes play a pivotal role in Gln synthesis via an astrocyte-specific enzyme, the GS24. Correspondently, the increased mI levels correlated with Gln levels in the ACC of individuals with autism. This might reflect that reactive astrocytes might produce higher Gln amounts in the brain of adults with autism; however, MRS-measured Gln levels may not solely reflect Gln in astrocytes. The disrupted homeostasis of the Glu-Gln cycle caused by astrocyte-specific Glu transporter (GLT1) deficiency was reported to induce pathological repetitive behaviors; this autistic-like behavior was alleviated by NMDA receptor inhibitors46. In addition, a human study reported increased GS plasma levels in individuals with autism47. These studies suggest that the disrupted homeostasis of the Glu-Gln cycle induced by reactive astrocytes might be associated with the pathology of autism. In the present study, we also observed a correlation between AQ Attention switching subscale scores and Gln and mI levels in the ACC, where neural activity is associated with attentional control48. Collectively, reactive astrocytes might be associated with increased excitatory balance, which may have potential applications for the development of drug treatments for autism49.

On the other hand, we found that both adults with autism and TD subjects exhibited a negative correlation of DA D1R bindings with Gln levels in the ACC. Our findings indicated that DA D1R activation was involved in Gln synthesis (Fig. 5). Some animal studies suggested a regulatory role of dopaminergic neurons in Gln synthesis28,50,51. Rodrigues et al.28 reported that blocking DA D1R or DA depletion in mice augmented Gln levels. Another study documented that DA denervation by MPTP induced elevated Gln levels accompanied by increased GS activity50. Moreover, DA D1R agonists can inhibit the ionotropic glutamate receptor, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic51. We speculated the existence of physiological mechanisms of inhibition of Gln synthesis related to dopaminergic D1R signaling in the ACC of adults with autism and TD subjects. In the autistic brain, reactive astrocytes might intensify Gln synthesis independently of the inhibitory role of dopaminergic neurons.

Summary of the enhanced Glu-Gln metabolism in autism. In physiological conditions, astrocytes absorb the Glu released from presynaptic neurons, where it is synthesized to Gln by GS. Subsequently, the synthesized Gln is transferred to presynaptic neurons and converted to Glu. In this Glu-Gln cycle, we speculate that DA D1R signals control Gln synthesis from Glu. Moreover, the Glu-Gln metabolism might be enhanced due to astrocyte activation in the autistic brain. DA D1R, dopamine D1 receptor; Gln, glutamine; Glu, glutamate; GS, glutamine synthetase.

This study has some limitations. First, it focused on the ACC but did not investigate other brain regions, based on the assumption of its involvement in the pathology of autism. Further research is needed to comprehensively understand the relationship between autism and all brain regions. Second, some data could not withstand corrections for multiple comparisons because diverse metabolites and neurotransmission components were analyzed in comparative and correlational fashions. Therefore, our findings should be interpreted with caution regarding this point. Finally, we did not evaluate differences in adaptive functioning among the autistic disorder, Asperger syndrome, and TD groups. Such investigations would be performed in an additional study with an expanded sample size, given that individuals with Asperger syndrome exhibit impaired social adaptive skills.

In conclusion, the present study indicated an excessive E/I balance accompanied by increased Gln levels, associated with reactive astrocytes in the ACC of individuals with autism. Under the physiological inhibitory control of Gln synthesis by dopaminergic signaling, the enhanced Glu-Gln metabolism induced by reactive astrocytes might be one pathological factor responsible for the E/I imbalance found in autism. Our findings lend further evidence to the role of astroglial activity in Glu-Gln metabolism, its relationship with attention switching, and the inhibitory role of dopamine receptors in Gln synthesis.

The participants’ demographic profiles are summarized in Table 130. Briefly, we evaluated 18 adults with autism and 20 TD individuals with matching age, intelligence quotient (IQ), and handedness. All subjects underwent MRS and PET scans. All participants were adult males, nonsmokers, and not exposed to monoamine-acting drugs. The TD group ranged in age from 20 to 42 years, while the autism group ranged in age from 23 to 46 years. The participants were volunteers diagnosed with autism at the outpatient department of Showa University Karasuyama Hospital. Both participants and caregivers, having relevant information about participants’ early childhood, were interviewed for ~ 3 h by > 3 experienced psychiatrists and one clinical psychologist using the Diagnostic and Statistical Manual of Mental Disorders, fourth edition text revision (DSM-IV-TR) based on the following criteria: patient’s developmental history, illness presence, life history, and family history. The final diagnosis of autistic or Asperger syndrome was made based on the consensus among the psychiatrists and the clinical psychologist. The autism group consisted of 16 adults with autistic disorder and two with Asperger syndrome. No individual was diagnosed with a pervasive developmental disorder not otherwise specified (PDD-NOS) in the present study. According to the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID), none of the participants with autism met the diagnostic criteria for substance use disorder, bipolar disorder, or schizophrenia.

TD subjects were recruited from the general population via advertisements and word-of-mouth. According to the SCID, none of these individuals met the diagnostic criteria for any psychiatric disorders and had no family history of serious medical or surgical illness, neurodevelopmental diseases, substance abuse, or neurodevelopmental diseases.

This study was approved by the Radiation Drug Safety Committee and the Institutional Review Board of the former National Institutes for former Quantum and Radiological Science and Technology, Japan, and by the institutional review board of Showa University Karasuyama Hospital, Japan. This study was conducted in accordance with the World Medical Association Code of Ethics. Written informed consent was obtained from all participants after providing them with a complete description of the study.

Prior to the bioanalytical study, the IQ levels of adults with autism were evaluated by the Wechsler Adult Intelligence Scale-Third Edition (WAIS-III). Since IQ scores were > 75 points in all adults with autism, we considered they had high-functioning autism. There was no difference in IQ between the autistic disorder and Asperger syndrome groups (mean [SD], 102.8 [17.1] in the autistic disorder group versus 111.0 [7.1] in the Asperger syndrome group; t = − 0.65, p = 0.52) (Fig. S1). Meanwhile, the IQ levels of individuals with TD were assessed with the Japanese version of the National Adult Reading Test (JART), which is reported to predict the full IQ score scale52. We assessed the autistic traits of adults with autism using AQ36.

We performed all MRI and MRS tests using a 3 T scanner (Siemens MAGNETOM Verio, Erlangen, Germany) with a 32-channel receiving head coil. The 3D volumetric acquisition of a T1-weighted gradient-echo sequence produced a gapless series of thin sagittal sections (repetition time (TR) = 2,300 ms; TE = 1.95 ms; inversion time (TI) = 900 ms; field of view (FOV) = 250 mm; flip angle = 9°; acquisition matrix = 256 × 256; axial slices thickness = 1 mm). To determine the voxels of interest (VOIs) localized to the ACC, we used anatomical images (Fig. 1).

As previously described, we applied MRS on the ACC using a special sequence37 with the following parameters20: TE = 8.5 ms, TR = 3000 ms, average number = 128, and VOI dimensions = 30 × 20 × 20 mm3. After running a 3D shim (Syngo MR version for B17, Siemens, Erlangen, Germany), we performed a manual shim so that the line width of the water spectrum in magnitude mode was < 20 Hz. Prior to the SPECIAL localization sequence, we applied an outer volume and water suppression with variable-pulse power and optimized relaxation delays53. We computed the tissue composition within the VOI based on 3D T1-weighted image segmentation using Gannet3.054. We calculated the water concentration in the white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) for the spectral analysis using a linear combination model (LCModel)55 based on the volume fractions of these segments. The concentrations were 35,880 mM, 43,300 mM, and 55,556 mM, respectively. We obtained the signal-to-noise ratio (SNR) by dividing the peak height of N-acetyl aspartic acid (NAA) at 2.01 ppm by the standard deviation (SD) of noise. For all spectra, we performed the LCModel quantification in a spectral window of 0.2–4.2 ppm. Macromolecules (MM) were fitted using LCModel’s default parameterized MM resonances while using the default LCModel baseline parameters.

We applied a weighted combination of receiver channels, followed by motion corruption average removal, spectrum registration for frequency and phase drift correction, and pre-subtraction sub-spectral alignment in MATLAB 2019a (The Mathworks, Natick, MA, USA) using the FID-A toolkit before signal averaging and data analysis56. To analyze MRS data, we used the LCModel with a basis set of simulated spectra comprising the following 20 metabolites: Aspartate (Asp), phosphocholine (PCh), creatine (Cr), phosphocreatine (PCr), GABA, Gln, Glu, glutathione (GSH), glycine (Gly), mI, lactate (Lac), NAA, scyllo-inositol (Scyllo), taurine (Tau), glucose (Glc), N-acetylaspartylglutamate (NAAG), glycerophosphocholine (GPC), phosphorylethanolamine (PE), Serine (Ser), and MM (Fig. 1). Among these metabolites, we focused on GABA, Gln, and Glu to evaluate the E/I ratio and on mI as an astrocyte marker43,57. The spectral SNR and the linewidth (LCModel output) in the ACC for the subjects with autism and TD were 76.6 ± 9.1 and 0.028 ± 0.004 ppm, respectively. We corrected the metabolite levels by segmenting GM, WM, and CSF in the VOI58.

We used the same procedures as in our previous study regarding PET data30. Each participant underwent a PET scan with 11C-SCH2339059 to visualize DA D1Rs. All PET scans were carried out with a Biograph mCT flow system (Siemens Healthcare, Erlangen Germany) which provides 109 transaxial sections with an axial FOV of 21.8 cm. The intrinsic in-plane and axial spatial resolutions yielded by this scanner are 5.9 and 5.5 mm full-width at half-maximum. Prior to the emission data, we conducted a computed tomography (CT) scan for attenuation correction. To minimize participants’ head movement during PET measurements, we used a head fixation device. However, one person with autism was excluded from PET analysis due to significant intra-frame motion. In addition, due to fatigue complaints, the 11C-SCH23390 PET scan was terminated at 52 min after radiotracer injection in one subject with TD.

We acquired a list-mode data with a PET-CT system for 60 min immediately after intravenous rapid bolus injection of 11C-SCH23390. Then, we sorted and rebinned the data with 38-frame signatures increasing from 20 s to 4 min (20 s × 12, 1 min × 16, and 4 min × 10). We corrected the sinograms for attenuation using the CT images, using the delayed coincidence counting method for random samples and the single-scatter simulation method for scatter samples. Subsequently, we reconstructed the modified sinograms using a filtered back-projection algorithm with a Hanning filter (4.0 mm full-width at half-maximum).

We performed co-registration of motion-corrected PET images with the corresponding individual’s T1-weighted MRI using PMOD® software ver. 3.8 (PMOD Technologies Ltd, Zurich, Switzerland). We used FreeSurfer software (version 6.0.0; http://surfer.nmr.harvard.edu) for surface-based cortical reconstruction and volumetric subcortical segmentation for each T1-weighted image. We defined regions of interest (ROIs) using brain atlases60,61,62. 11C-SCH23390 binding to DA D1Rs was quantified as nondisplaced tissue (BPND). Next, we calculated the BPND for the target ROI in the ACC using a three-parameter simplified reference tissue model63 with the cerebellar cortex excluding the vermis as a reference tissue. ROIs were found to be similar to our previous study30.

The results are indicated as mean and SD (mean [SD]). We applied independent sample t-tests and χ2-tests for the statistical examination for all participants regarding differences in demographics, voxel-wise MRS data, AQ scores, and regional PET radioligand binding. To test correlations between MRS data, AQ subscale scores, and PET radioligand binding, we performed Pearson correlation and Spearman's partial rank-order correlation analyses. The latter was applied for AQ communication and imagination subscale scores, since they are not normally distributed. Statistical analyses were conducted with IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, N.Y., USA). The statistical significance threshold was set at p < 0.05 (two-tailed) for group comparisons and correlation analyses.

Data can be obtained upon reasonable request. Anonymized raw data supporting the findings of the present study may be shared upon request with the corresponding author.

Richa, S., Fahed, M., Khoury, E. & Mishara, B. Suicide in autism spectrum disorders. Arch. Suicide Res. 18, 327–339. https://doi.org/10.1080/13811118.2013.824834 (2014).

Article PubMed Google Scholar

Mason, D. et al. Predictors of quality of life for autistic adults. Autism Res. 11, 1138–1147. https://doi.org/10.1002/aur.1965 (2018).

Article PubMed PubMed Central Google Scholar

Christensen, D. L. et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years – autism and developmental disabilities monitoring network, 11 sites, United States, 2012. M.M.W.R Surveill. Summ. 65, 1–23. https://doi.org/10.15585/mmwr.ss6513a1 (2018).

Article Google Scholar

McCracken, J. T. et al. Drug development for autism spectrum disorder (ASD): Progress, challenges, and future directions. Eur. Neuropsychopharmacol. 48, 3–31. https://doi.org/10.1016/j.euroneuro.2021.05.010 (2021).

Article CAS PubMed PubMed Central Google Scholar

Kourosh-Arami, M., Hosseini, N. & Komaki, A. Brain is modulated by neuronal plasticity during postnatal development. J. Physiol. Sci. 71, 34. https://doi.org/10.1186/s12576-021-00819-9 (2021).

Article PubMed Google Scholar

Port, R. G., Oberman, L. M. & Roberts, T. P. Revisiting the excitation/inhibition imbalance hypothesis of ASD through a clinical lens. Br. J. Radiol. 92, 20180944. https://doi.org/10.1259/bjr.20180944 (2019).

Article PubMed PubMed Central Google Scholar

Rubenstein, J. L. & Merzenich, M. M. Model of autism: Increased ratio of excitation/inhibition in key neural systems. Genes Brain Behav. 2, 255–267. https://doi.org/10.1034/j.1601-183x.2003.00037.x (2003).

Article CAS PubMed Google Scholar

Takarae, Y. & Sweeney, J. Neural hyperexcitability in autism spectrum disorders. Brain Sci. 7, 129. https://doi.org/10.3390/brainsci7100129 (2017).

Article CAS PubMed PubMed Central Google Scholar

Antoine, M. W., Langberg, T., Schnepel, P. & Feldman, D. E. Increased excitation-inhibition ratio stabilizes synapse and circuit excitability in four autism mouse models. Neuron 101, 648-661.e4. https://doi.org/10.1016/j.neuron.2018.12.026 (2019).

Article CAS PubMed PubMed Central Google Scholar

Gkogkas, C. G. et al. Autism-related deficits via dysregulated eIF4E-dependent translational control. Nature 493, 371–377. https://doi.org/10.1038/nature11628 (2013).

Article ADS CAS PubMed Google Scholar

Yizhar, O. et al. Neocortical excitation/inhibition balance in information processing and social dysfunction. Nature 477, 171–178. https://doi.org/10.1038/nature10360 (2011).

Article ADS CAS PubMed PubMed Central Google Scholar

Milovanovic, M. & Grujicic, R. Electroencephalography in assessment of autism spectrum disorders: A review. Front. Psychiatry 12, 686021. https://doi.org/10.3389/fpsyt.2021.686021 (2021).

Article PubMed PubMed Central Google Scholar

Ford, T. C. & Crewther, D. P. A comprehensive review of the 1H-MRS metabolite spectrum in autism spectrum disorder. Front. Mol. Neurosci. 9, 14. https://doi.org/10.3389/fnmol.2016.00014 (2016).

Article CAS PubMed PubMed Central Google Scholar

Takado, Y. et al. MRS-measured glutamate versus GABA reflects excitatory versus inhibitory neural activities in awake mice. J. Cereb. Blood Flow Metab. 42, 197–212. https://doi.org/10.1177/0271678X211045449 (2022).

Article CAS PubMed Google Scholar

Ajram, L. A. et al. The contribution of [1H] magnetic resonance spectroscopy to the study of excitation-inhibition in autism. Prog. Neuropsychopharmacol. Biol. Psychiatry 89, 236–244. https://doi.org/10.1016/j.pnpbp.2018.09.010 (2019).

Article CAS PubMed Google Scholar

Cochran, D. M. et al. Relationship among glutamine, γ-aminobutyric acid, and social cognition in autism spectrum disorders. J. Child Adolesc. Psychopharmacol. 25, 314–322. https://doi.org/10.1089/cap.2014.0112 (2015).

Article CAS PubMed PubMed Central Google Scholar

Horder, J. et al. Glutamate and GABA in autism spectrum disorder-a translational magnetic resonance spectroscopy study in man and rodent models. Transl. Psychiatry 8, 106. https://doi.org/10.1038/s41398-018-0155-1 (2018).

Article CAS PubMed PubMed Central Google Scholar

van Tebartz Elst, L. et al. Disturbed cingulate glutamate metabolism in adults with high-functioning autism spectrum disorder: Evidence in support of the excitatory/inhibitory imbalance hypothesis. Mol. Psychiatry 19, 1314–1325. https://doi.org/10.1038/mp.2014.62 (2014).

Article CAS Google Scholar

Öz, G. et al. Advanced single voxel 1H magnetic resonance spectroscopy techniques in humans: Experts’ consensus recommendations. N.M.R. Biomed. 34, e4236. https://doi.org/10.1002/nbm.4236 (2020).

Article MathSciNet CAS Google Scholar

Takado, Y. et al. Association between brain and plasma glutamine levels in healthy young subjects investigated by MRS and LC/MS. Nutrients 11, 1649. https://doi.org/10.3390/nu11071649 (2019).

Article CAS PubMed PubMed Central Google Scholar

Bonansco, C. & Fuenzalida, M. Plasticity of hippocampal excitatory-inhibitory balance: Missing the synaptic control in the epileptic brain. Neural Plast. 2016, 8607038. https://doi.org/10.1155/2016/8607038 (2016).

Article CAS PubMed PubMed Central Google Scholar

Henstridge, C. M., Tzioras, M. & Paolicelli, R. C. Glial contribution to excitatory and inhibitory synapse loss in neurodegeneration. Front. Cell. Neurosci. 13, 63. https://doi.org/10.3389/fncel.2019.00063 (2019).

Article CAS PubMed PubMed Central Google Scholar

Nagai, J. et al. Behaviorally consequential astrocytic regulation of neural circuits. Neuron 109, 576–596. https://doi.org/10.1016/j.neuron.2020.12.008 (2021).

Article CAS PubMed Google Scholar

Rose, C. F., Verkhratsky, A. & Parpura, V. Astrocyte glutamine synthetase: Pivotal in health and disease. Biochem. Soc. Trans. 41, 1518–1524. https://doi.org/10.1042/BST20130237 (2013).

Article CAS PubMed Google Scholar

Gzielo, K. & Nikiforuk, A. Astroglia in autism spectrum disorder. Int. J. Mol. Sci. 22, 11544. https://doi.org/10.3390/ijms222111544 (2021).

Article CAS PubMed PubMed Central Google Scholar

Fatemi, S. H., Folsom, T. D., Reutiman, T. J. & Lee, S. Expression of astrocytic markers aquaporin 4 and connexin 43 is altered in brains of subjects with autism. Synapse 62, 501–507. https://doi.org/10.1002/syn.20519 (2008).

Article CAS PubMed PubMed Central Google Scholar

Wang, J., Zou, Q., Han, R., Li, Y. & Wang, Y. Serum levels of glial fibrillary acidic protein in Chinese children with autism spectrum disorders. Int. J. Dev. Neurosci. 57, 41–45. https://doi.org/10.1016/j.ijdevneu.2017.01.004 (2017).

Article CAS PubMed Google Scholar

Rodrigues, T. B., Granado, N., Ortiz, O., Cerdán, S. & Moratalla, R. Metabolic interactions between glutamatergic and dopaminergic neurotransmitter systems are mediated through D1 dopamine receptors. J. Neurosci. Res. 85, 3284–3293. https://doi.org/10.1002/jnr.21302 (2007).

Article CAS PubMed Google Scholar

Pavăl, D. & Micluția, I. V. The dopamine hypothesis of autism spectrum disorder revisited: Current status and future prospects. Dev. Neurosci. 43, 73–83. https://doi.org/10.1159/000515751 (2021).

Article CAS PubMed Google Scholar

Kubota, M. et al. Binding of dopamine D1 receptor and noradrenaline transporter in individuals with autism spectrum disorder: A PET study. Cereb. Cortex 30, 6458–6468. https://doi.org/10.1093/cercor/bhaa211 (2020).

Article PubMed Google Scholar

Uppal, N. et al. Neuropathology of the anterior midcingulate cortex in young children with autism. J. Neuropathol. Exp. Neurol. 73, 891–902. https://doi.org/10.1097/nen.0000000000000108 (2014).

Article PubMed Google Scholar

Zikopoulos, B. & Barbas, H. Changes in prefrontal axons may disrupt the network in autism. J. Neurosci. 30, 14595–14609. https://doi.org/10.1523/jneurosci.2257-10.2010 (2010).

Article CAS PubMed PubMed Central Google Scholar

Dichter, G. S. et al. Reward circuitry function in autism spectrum disorders. Soc. Cogn. Affect. Neurosci. 7, 160–172. https://doi.org/10.1093/scan/nsq095 (2012).

Article PubMed Google Scholar

Balsters, J. H., Mantini, D., Apps, M. A. J., Eickhoff, S. B. & Wenderoth, N. Connectivity-based parcellation increases network detection sensitivity in resting state fMRI: An investigation into the cingulate cortex in autism. NeuroImage Clin. 11, 494–507. https://doi.org/10.1016/j.nicl.2016.03.016 (2016).

Article PubMed PubMed Central Google Scholar

Trakoshis, S. et al. Intrinsic excitation-inhibition imbalance affects medial prefrontal cortex differently in autistic men versus women. Elife https://doi.org/10.7554/eLife.55684 (2020).

Article PubMed PubMed Central Google Scholar

Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J. & Clubley, E. The autism-spectrum quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. J. Autism Dev. Disord. 31, 5–17. https://doi.org/10.1023/a:1005653411471 (2001).

Article CAS PubMed Google Scholar

Mekle, R. et al. MR spectroscopy of the human brain with enhanced signal intensity at ultrashort echo times on a clinical platform at 3T and 7T. Magn. Reson. Med. 61, 1279–1285. https://doi.org/10.1002/mrm.21961 (2009).

Article CAS PubMed Google Scholar

Budimirovic, D. B. & Kaufmann, W. E. What can we learn about autism from studying fragile X syndrome?. Dev Neurosci 33, 379–394. https://doi.org/10.1159/000330213 (2011).

Article CAS PubMed PubMed Central Google Scholar

Duy, P. Q. & Budimirovic, D. B. Fragile X syndrome: Lessons learned from the most translated neurodevelopmental disorder in clinical trials. Transl Neurosci 8, 7–8. https://doi.org/10.1515/tnsci-2017-0002 (2017).

Article CAS PubMed PubMed Central Google Scholar

Budimirovic, D. B. et al. Updated report on tools to measure outcomes of clinical trials in fragile X syndrome. J Neurodev Disord 9, 14. https://doi.org/10.1186/s11689-017-9193-x (2017).

Article PubMed PubMed Central Google Scholar

Gabis, L. et al. 1H-magnetic resonance spectroscopy markers of cognitive and language ability in clinical subtypes of autism spectrum disorders. J. Child Neurol. 23, 766–774. https://doi.org/10.1177/0883073808315423 (2008).

Article PubMed Google Scholar

Vasconcelos, M. M. et al. Proton magnetic resonance spectroscopy in school-aged autistic children. J. Neuroimaging 18, 288–295. https://doi.org/10.1111/j.1552-6569.2007.00200.x (2008).

Article PubMed Google Scholar

Ross, B. & Bluml, S. Magnetic resonance spectroscopy of the human brain. Anat. Rec. 265, 54–84. https://doi.org/10.1002/ar.1058 (2001).

Article CAS PubMed Google Scholar

Haida, O. et al. Sex-dependent behavioral deficits and neuropathology in a maternal immune activation model of autism. Transl. Psychiatry 9, 124. https://doi.org/10.1038/s41398-019-0457-y (2019).

Article PubMed PubMed Central Google Scholar

Matta, S. M., Moore, Z., Walker, F. R., Hill-Yardin, E. L. & Crack, P. J. An altered glial phenotype in the NL3R451C mouse model of autism. Sci. Rep. 10, 14492. https://doi.org/10.1038/s41598-020-71171-y (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Aida, T. et al. Astroglial glutamate transporter deficiency increases synaptic excitability and leads to pathological repetitive behaviors in mice. Neuropsychopharmacology 40, 1569–1579. https://doi.org/10.1038/npp.2015.26 (2015).

Article CAS PubMed PubMed Central Google Scholar

Hamed, N. O. et al. Understanding the roles of glutamine synthetase, glutaminase, and glutamate decarboxylase autoantibodies in imbalanced excitatory/inhibitory neurotransmission as etiological mechanisms of autism. Psychiatry Clin. Neurosci. 72, 362–373. https://doi.org/10.1111/pcn.12639 (2018).

Article CAS PubMed Google Scholar

Totah, N. K., Kim, Y. B., Homayoun, H. & Moghaddam, B. Anterior cingulate neurons represent errors and preparatory attention within the same behavioral sequence. J. Neurosci. 29, 6418–6426. https://doi.org/10.1523/jneurosci.1142-09.2009 (2009).

Article CAS PubMed PubMed Central Google Scholar

Montanari, M., Martella, G., Bonsi, P. & Meringolo, M. Autism spectrum disorder: Focus on glutamatergic neurotransmission. Int. J. Mol. Sci. 23, 3861. https://doi.org/10.3390/ijms23073861 (2022).

Article CAS PubMed PubMed Central Google Scholar

Chassain, C. et al. Does MPTP intoxication in mice induce metabolite changes in the nucleus accumbens? A 1H nuclear MRS study. N.M.R. Biomed. 26, 336–347. https://doi.org/10.1002/nbm.2853 (2013).

Article CAS Google Scholar

Darvish-Ghane, S., Quintana, C., Beaulieu, J. M. & Martin, L. J. D1 receptors in the anterior cingulate cortex modulate basal mechanical sensitivity threshold and glutamatergic synaptic transmission. Mol. Brain 13, 121. https://doi.org/10.1186/s13041-020-00661-x (2020).

Article CAS PubMed PubMed Central Google Scholar

Matsuoka, K., Uno, M., Kasai, K., Koyama, K. & Kim, Y. Estimation of premorbid IQ in individuals with Alzheimer’s disease using Japanese ideographic script (Kanji) compound words, Japanese version of National Adult Reading Test. Psychiatry Clin Neurosci 60, 332–339. https://doi.org/10.1111/j.1440-1819.2006.01510.x (2006).

Article PubMed Google Scholar

Tkác, I. et al. In vivo 1H NMR spectroscopy of the human brain at 7 T. Magn. Reson. Med. 46, 451–456. https://doi.org/10.1002/mrm.1213 (2001).

Article PubMed Google Scholar

Harris, A. D., Puts, N. A. & Edden, R. A. Tissue correction for GABA-edited MRS: Considerations of voxel composition, tissue segmentation, and tissue relaxations. J. Magn. Reson. Imaging 42, 1431–1440. https://doi.org/10.1002/jmri.24903 (2015).

Article PubMed PubMed Central Google Scholar

Provencher, S. W. Estimation of metabolite concentrations from localized in vivo proton NMR spectra. Magn. Reson. Med. 30, 672–679. https://doi.org/10.1002/mrm.1910300604 (1993).

Article CAS PubMed Google Scholar

Simpson, R., Devenyi, G. A., Jezzard, P., Hennessy, T. J. & Near, J. Advanced processing and simulation of MRS data using the FID appliance (FID-A)-an open source MATLAB-based toolkit. Magn. Reson. Med. 77, 23–33. https://doi.org/10.1002/mrm.26091 (2017).

Article CAS PubMed Google Scholar

Pouwels, P. J. & Frahm, J. Regional metabolite concentrations in human brain as determined by quantitative localized proton MRS. Magn. Reson. Med. 39, 53–60. https://doi.org/10.1002/mrm.1910390110 (1998).

Article CAS PubMed Google Scholar

Panchal, H. et al. Neuro-metabolite changes in a single season of university ice hockey using magnetic resonance spectroscopy. Front. Neurol. 9, 616. https://doi.org/10.3389/fneur.2018.00616 (2018).

Article PubMed PubMed Central Google Scholar

Farde, L., Halldin, C., Stone-Elander, S. & Sedvall, G. PET analysis of human dopamine receptor subtypes using 11C-SCH 23390 and 11C-raclopride. Psychopharmacol. (Berl.) 92, 278–284. https://doi.org/10.1007/BF00210831 (1987).

Article CAS Google Scholar

Fischl, B. et al. Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron 33, 341–355. https://doi.org/10.1016/s0896-6273(02)00569-x (2002).

Article CAS PubMed Google Scholar

Greve, D. N. et al. Different partial volume correction methods lead to different conclusions: An (18)F-FDG-PET study of aging. Neuroimage 132, 334–343. https://doi.org/10.1016/j.neuroimage.2016.02.042 (2016).

Article PubMed Google Scholar

Klein, A. & Tourville, J. 101 labeled brain images and a consistent human cortical labeling protocol. Front. Neurosci. 6, 171. https://doi.org/10.3389/fnins.2012.00171 (2012).

Article PubMed PubMed Central Google Scholar

Lammertsma, A. A. & Hume, S. P. Simplified reference tissue model for PET receptor studies. Neuroimage 4, 153–158. https://doi.org/10.1006/nimg.1996.0066 (1996).

Article CAS PubMed Google Scholar

Download references

We thank the staff of the clinical research section for their assistance as clinical coordinators, the staff of the Department of Molecular Imaging and Theranostics for their support with the MRI scans, the PET and the MRI operators particularly Takasama Maeda for their imaging scans, the staff of the Department of Radiopharmaceuticals Development for their radioligand synthesis, Atsuo Waki and his team for the quality assurance of the radioligands, and Takashi Horiguchi for his assistance as research administrator, Jamie Near for his assistance in MRS protocol of SPECIAL sequence. We also wish to extend our gratitude to the research team of the Medical Institute of Developmental Disabilities Research at Showa University for their assistance in data acquisition.

This research was funded by Grants from the Joint Usage/Research Program of the Medical Institute of Developmental Disabilities Research, Showa University; SENSHIN Medical Research Foundation; the Japan Society for the Promotion of Science (Grant-in-Aid for Young Scientists, 19K17101 to MK); MEXT KAKENHI Grant numbers 19H01041; and the Japan Agency for Medical Research and Development (the program for Brain/MINDS-beyond, 22dm0307105h0304, 22dk0207063h0001).

Department of Functional Brain Imaging, Institute for Quantum Medical Science, NationalInstitutes for Quantum Science and Technology (QST), 4-9-1 Anagawa, Inage-ku, Chiba-shi, Chiba, 263-8555, Japan

Masaki Oya, Kiwamu Matsuoka, Manabu Kubota, Keisuke Takahata, Kenji Tagai, Yasuharu Yamamoto, Hitoshi Shimada, Chie Seki, Tetsuya Suhara, Yuhei Takado & Makoto Higuchi

Department of Psychiatry and Behavioral Sciences, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan

Masaki Oya, Junya Fujino, Genichi Sugihara & Hidehiko Takahashi

Department of Psychiatry, Nara Medical University, Kashihara-shi, Nara, Japan

Kiwamu Matsuoka

Department of Psychiatry, Graduate School of Medicine, Kyoto University, Kyoto-shi, Kyoto, Japan

Manabu Kubota & Shisei Tei

Medical Institute of Developmental Disabilities Research, Showa University, Setagaya-ku, Tokyo, Japan

Manabu Kubota, Junya Fujino, Shisei Tei, Takashi Itahashi, Yuta Y. Aoki, Haruhisa Ohta, Ryu-ichiro Hashimoto, Motoaki Nakamura & Nobumasa Kato

Institute of Applied Brain Sciences, Waseda University, Tokorozawa-shi, Saitama, Japan

Shisei Tei

School of Human and Social Sciences, Tokyo International University, Kawagoe-shi, Saitama, Japan

Shisei Tei

Department of Neuropsychiatry, Keio University School of Medicine, Shinjuku-ku, Tokyo, Japan

Keisuke Takahata & Yasuharu Yamamoto

Center for Integrated Human Brain Science, Brain Research Institute, Niigata University, Niigata-shi, Niigata, Japan

Hitoshi Shimada

Department of Psychiatry, School of Medicine, Showa University, Setagaya-ku, Tokyo, Japan

Haruhisa Ohta

Department of Language Sciences, Graduate School of Humanities, Tokyo Metropolitan University, Hachioji-shi, Tokyo, Japan

Ryu-ichiro Hashimoto

Department of Molecular Imaging and Theranostics, Institute for Quantum Medical Science, National Institutes for Quantum Science and Technology, Chiba-shi, Chiba, Japan

Takayuki Obata

Department of Advanced Nuclear Medicine Sciences, Institute for Quantum Medical Science, National Institutes for Quantum Science and Technology, Chiba-shi, Chiba, Japan

Ming-Rong Zhang

Kanagawa Psychiatric Center, Yokohama-shi, Kanagawa, Japan

Motoaki Nakamura

Center for Brain Integration Research, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan

Hidehiko Takahashi

You can also search for this author in PubMed Google Scholar

M.O., K.M., M.K., J.F, M.N., N.K., H.T. and M.H were involved in the conception of the study, M.K., J.F., Keisuke.T., Kenji.T., Y.Y., H.S., C.S., T.I., Y.Y.A., H.O. and R.H. were involved in the acquisition of the data, and M.O., K.M., S.T., , G.S., T.O., M.Z., T.S., Y.T., H.T., and M.H in the interpretation of the data. M.O. has drafted the manuscript, and all authors have revised it. All authors have given final approval for the version to be published.

Correspondence to Kiwamu Matsuoka.

The authors declare no competing interests.

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

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 changes were made. 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/4.0/.

Reprints and Permissions

Oya, M., Matsuoka, K., Kubota, M. et al. Increased glutamate and glutamine levels and their relationship to astrocytes and dopaminergic transmissions in the brains of adults with autism. Sci Rep 13, 11655 (2023). https://doi.org/10.1038/s41598-023-38306-3

Download citation

Received: 11 January 2023

Accepted: 06 July 2023

Published: 19 July 2023

DOI: https://doi.org/10.1038/s41598-023-38306-3

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.