Hostname: page-component-7bb8b95d7b-lvwk9 Total loading time: 0 Render date: 2024-10-06T23:55:28.527Z Has data issue: false hasContentIssue false

Glutamine + glutamate level predicts the magnitude of microstructural organization in the gray matter in the healthy elderly

Published online by Cambridge University Press:  03 October 2019

Tomokazu Motegi*
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
Kosuke Narita
Affiliation:
Department of Psychiatry, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan
Kazuyuki Fujihara
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
Masato Kasagi
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
Yusuke Suzuki
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
Minami Tagawa
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
Koichi Ujita
Affiliation:
Department of Diagnostic Radiology and Nuclear Medicine, Gunma University Graduate School of Medicine, Maebashi, Gunma 371-8511, Japan
Jamie Near
Affiliation:
Douglas Mental Health University Institute and Department of Psychiatry, McGill University, Montreal, QC, Canada
Masato Fukuda
Affiliation:
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan
*
Correspondence should be addressed to: Tomokazu Motegi, Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511, Japan. Phone: +81 27 220 8185; Fax: +81 27 220 8187. Email address: motegi@atr.jp (T. Motegi).
Rights & Permissions [Opens in a new window]

Abstract

Background:

Diffusion tensor imaging (DTI), which is a technique for measuring the degree and direction of movement of water molecules in tissue, has been widely used to noninvasively assess white matter (WM) or gray matter (GM) microstructures in vivo. Mean diffusivity (MD), which is the average diffusion across all directions, has been considered as a marker of WM tract degeneration or extracellular space enlargement in GM. Recent lines of evidence suggest that cortical MD can better identify early-stage Alzheimer’s disease than structural morphometric parameters in magnetic resonance imaging. However, knowledge of the relationships between cortical MD and other biological factors in the same cortical region, e.g. metabolites, is still limited.

Methods:

Thirty-three healthy elderly individuals [aged 50–77 years (mean, 63.8±7.4 years); 11 males and 22 females] were enrolled. We estimated the associations between cortical MD and neurotransmitter levels. Specifically, we measured levels of γ-aminobutyric acid (GABA) and glutamate + glutamine (Glx), which are inhibitory and excitatory neurotransmitters, respectively, in medial prefrontal cortex (mPFC) and posterior cingulate cortex (PCC) using MEGA-PRESS magnetic resonance spectroscopy, and we measured regional cortical MD using DTI.

Results:

Cortical MD was significantly negatively associated with Glx levels in both mPFC and PCC. No significant association was observed between cortical MD and GABA levels in either GM region.

Conclusion:

Our findings suggest that degeneration of microstructural organization in GM, as determined on the basis of cortical MD measured by DTI, is accompanied by the decline of Glx metabolism within the same GM region.

Type
Original Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© International Psychogeriatric Association 2019

Introduction

Over the last 30 years, various methods of magnetic resonance imaging (MRI) have been developed and used to assess brain alterations associated with normal aging and aged-related neurodegenerative diseases, such as mild cognitive impairment and Alzheimer’s disease (Elman et al., Reference Elman2017). Diffusion tensor imaging (DTI), a reliable MRI technique for measuring the degree and direction of movement of water molecules in brain tissue, has been widely used to noninvasively assess white matter (WM) or gray matter (GM) microstructures in vivo (Le Bihan et al., Reference Le Bihan, Breton, Lallemand, Grenier, Cabanis and Laval-Jeantet1986). Fractional anisotropy (FA), which is thought to be an index of WM integrity calculated by DTI, can reflect the state of neural fibers on the basis of the density, diameter, or coherence of axons (Le Bihan et al., Reference Le Bihan2001). Mean diffusivity (MD), which is the average magnitude of diffusion across all directions, has been considered as a marker of WM tract degeneration or extracellular space enlargement in GM (Elman et al., Reference Elman2017; Neil et al., Reference Neil, Miller, Mukherjee and Huppi2002).

Many previous DTI studies have revealed altered FA and MD even in normal aging (Abe et al., Reference Abe2008; Benedetti et al., Reference Benedetti2006; Garcia-Lazaro et al., Reference Garcia-Lazaro, Becerra-Laparra, Cortez-Conradis and Roldan-Valadez2016), age-related neurodegenerative diseases, such as mild cognitive impairment and Alzheimer’s disease (Nesteruk et al., Reference Nesteruk, Nesteruk, Styczynska, Barcikowska-Kotowicz and Walecki2016; Nishioka et al., Reference Nishioka, Poh and Sun2015), demyelinating diseases, such as multiple sclerosis (de Kouchkovsky et al., Reference de Kouchkovsky, Fieremans, Fleysher, Herbert, Grossman and Inglese2016), and neuropsychiatric diseases, such as depression and schizophrenia (Jiang et al., Reference Jiang2017; Singh et al., Reference Singh2016). Most of these studies have focused on WM or deep GM such as the hippocampus, because these brain regions have high directionality in water diffusion (Manna et al., Reference Manna, Piras, Caltagirone, Bossu, Sensi and Spalletta2015; Ziyan and Westin, Reference Ziyan and Westin2008), whereas water diffusion in the cerebral cortex has an isotropic direction at the level of conventional DTI resolution (Elman et al., Reference Elman2017).

On the other hand, recent lines of DTI evidence have shown that altered cortical MD may reflect the magnitude of microstructural organization in GM, that is, patients with mild cognitive impairment or Alzheimer’s disease have been reported to show increased cortical MD, compared with healthy controls in several GM regions including the posterior cingulate cortex (PCC), entorhinal cortex, amygdala, parahippocampal gyrus, middle temporal gyrus, superior and middle frontal gyri and bilateral supramarginal gyri (Lin et al., Reference Lin2016; Ray et al., Reference Ray2006; Rose et al., Reference Rose, Janke and Chalk2008; Weston et al., Reference Weston, Simpson, Ryan, Ourselin and Fox2015). Furthermore, cortical MD begins to increase more generally in middle age (Ni et al., Reference Ni, Chen, Liu, Huang, Shen and Chen2010). Noteworthy, a recent twin study showed that cortical MD is associated with genetic factors, distinct from cortical thickness or MD in WM (Elman et al., Reference Elman2017), suggesting that cortical MD measurement by DTI may have greater potential use for assessing degeneration of microstructural organization in the GM associated with normal aging or age-related diseases than morphometric GM volume estimation in T1-weighted imaging. However, knowledge of relationships between MD and metabolite levels in the same cortical regions is still limited.

In this study of healthy elderly people without dementia, we examined the associations between cortical MD and the levels of γ-aminobutyric acid (GABA) and glutamine + glutamate (Glx), which are inhibitory and excitatory neurotransmitters, respectively, in the medial prefrontal cortex (mPFC) and PCC by DTI and magnetic resonance spectroscopy (MRS) using MEGA-PRESS (Bauer et al., Reference Bauer2013; Jocham et al., Reference Jocham, Hunt, Near and Behrens2012). Decreased levels of GABA and Glx in several cortical areas including mPFC and PCC have been reported to be associated with aging (Goryawala et al., Reference Goryawala, Sheriff and Maudsley2016; Grachev et al., Reference Grachev, Swarnkar, Szeverenyi, Ramachandran and Apkarian2001) and to be found in patients with mild cognitive impairment and Alzheimer’s disease (Antuono et al., Reference Antuono, Jones, Wang and Li2001; Huang et al., Reference Huang, Liu, Yin, Qian, Shrestha and Ni2016; Riese et al., Reference Riese2015), that is, mPFC or PCC has been recognized as one of the most important brain areas associated with memory loss or cognitive decline in elderly people. In addition, decreased Glx levels in certain GM regions have been suggested to be associated with neuronal dysfunction and loss in that GM region (Segovia et al., Reference Segovia, Porras, Del Arco and Mora2001). We hypothesized that cortical MD in mPFC and PCC correlates negatively with the levels of GABA and Glx in the same volume of interest (VOI).

Methods

Subjects

Thirty-three individuals [aged 50–77 years (mean, 63.8±7.4 years); 11 males and 22 females] were enrolled in this study (Table 1) on the basis of the following exclusion criteria: history of dementia, neurological or psychiatric illness, diabetes mellitus, chronic alcoholism, smoking, and obesity with a body mass index above 25. All subjects were right-handed, as assessed using the Edinburgh Handedness Inventory (Oldfield, Reference Oldfield1971), and provided their written informed consent. The study protocol was approved by the Ethics Committee of Gunma University.

Table 1. Demographic characteristics of study subjects

MoCA, the Montreal Cognitive Assessment; JART, Japanese version of National Adult Reading Test; A’, signal detection measures of accuracy, Mean ± SD.

Psychological measurements

The Montreal Cognitive Assessment (MoCA) was administered to all subjects for the screening of cognitive function (Ihara et al., Reference Ihara, Okamoto and Takahashi2013; Nasreddine et al., Reference Nasreddine2005); although several cut-off points have been proposed for MoCA, we used scores ≥ 20 to exclude subjects with dementia and cognitive impairment, in accordance with a previous report (Waldron-Perrine and Axelrod, Reference Waldron-Perrine and Axelrod2012). The Japanese version of the National Adult Reading Test was then conducted to estimate premorbid IQs (Matsuoka et al., Reference Matsuoka, Uno, Kasai, Koyama and Kim2006; Nelson, Reference Nelson1982). The Cambridge neuropsychological test automated battery [CANTAB (Morris et al., Reference Morris, Evenden, Sahakian, Robbins, Stahl, Iversen and Goodman1987); Cambridge Cognition Ltd., Cambridge, United Kingdom] was also administered to each subject of this study and consisted of the following: the Spatial Recognition Memory (SRM) test of visual spatial memory in a two-choice forced discrimination paradigm and the Rapid Visual Information Processing (RVIP) test, which is a visual continuous performance task using digits instead of letters.

Acquisition of MRS and DTI and T1-weighted anatomical imaging data

MRS acquisition

Edited GABA and Glx MR spectra were acquired using the MEGA-PRESS sequence (Mescher et al., Reference Mescher, Merkle, Kirsch, Garwood and Gruetter1998) with the following acquisition parameters: TR = 2400 ms; TE = 68 ms; number of averages = 512 for the mPFC and 256 for PCC. In both regions, the prescribed MRS VOIs was 30 × 20 × 20 mm3. Based on the chemical shift difference between the 3 ppm GABA resonance and the 3.75 ppm Glx resonance, the chemical shift displacement between GABA and Glx in the direction of the excitation pulse (bandwidth =3708 Hz, selective in the left–right direction), was 0.75 mm. In the directions of refocusing pulses (bandwidth = 1106 Hz, selective in the anterior–posterior direction and in the superior–inferior direction), the chemical shift displacement between GABA and Glx was 1.67 mm. The VOI in mPFC was defined as follows. After drawing “line a” exactly on the rostral margin of the corpus callosum as the perpendicular axis to the anterior commissure–posterior commissure (AC–PC) line (see Fig. 1A), the VOI in mPFC was set along “line a” and on the inferoposterior corner located at the rostral edge of the genu. The VOI in PCC was set above the superior surface of the corpus callosum together with the diagonal line of the VOI aligned along “line b,” which is the perpendicular axis across the posterior edge of the splenium to the AC–PC line (see Fig. 1A).

Figure 1. Magnetic resonance spectroscopy (MRS) using MEGA-PRESS. The volumes of interest (30 × 20 × 20 mm3) in MRS, which were located on mPFC and PCC, are shown in A; “line a” is set exactly on the rostral margin of the corpus callosum as the perpendicular axis to the anterior commissure–posterior commissure (AC–PC) line. “line b” is drawn as the perpendicular axis through the anterior commissure to the AC–PC line. The edited spectrum (i.e. black line) and fitted curve (i.e. red line), which were obtained for the signal quantification of GABA, Glx, and Cr levels, are shown in B.

We measured the full-width at half-maximum (FWHM) of N-acetyl aspartate (NAA) peaks in the MEGA-PRESS spectra to determine the quality of shimming. The means ± standard deviations (SDs) were 6.864 ± 2.272 Hz in mPFC and 5.201± 2.203 Hz in PCC. We excluded the samples whose FWHM values of NAA peaks were large (i.e. FWHM > mean ± 2 SD). Although data from PCC were successfully collected from all the participants, those from mPFC of one participant were excluded from the analysis because the FWHM of the NAA peak was larger than the cut-off value. In addition, we evaluated motion artifacts by visual inspection of superimposed spectra for all the excitations during scanning using MEGA-PRESS. As a result, two samples obtained from mPFC were also excluded owing to motion artifacts.

DTI and T1-weighted anatomical imaging data acquisitions

DTI data of the participants were acquired using a clinical 3.0 Tesla Prisma MRI scanner (Siemens, Erlangen, Germany) with a 12-channel head coil. The entire brain was scanned by echo-planar imaging with 30 noncolinear motion probing gradients at a b-value of 1000 sec/mm2 and 5 T2-weighted b = 0 images. The acquisition parameters for the DTI scan were as follows: TE = 84 ms, TR = 6500 ms, 50 axial slices, slice thickness = 3 mm, field of view = 23 × 23 cm2, and matrix size = 128 × 128 (i.e. voxel size = 1.8 × 1.8 × 3.0 mm3). To improve signal-to-noise ratio, the scanning was repeated twice [number of excitation (NEX) = 2].

We also obtained high-resolution T1-weighted images for GM morphometry. MP-RAGE sequencing yielding 176 continuous slices of 1.0 mm thickness was carried out along the sagittal plane. The acquisition parameters were as follows: echo time = 2 ms; repetition time = 2000 ms; inversion time = 990 ms; flip angle = 9°; field of view = 256 mm; matrix size = 256 × 256; voxel size = 1 × 1 × 1 mm3.

Processing and analysis of MRS and DTI data

Measurement of Gaba and GLX Levels by MRS

Prior to signal averaging, each scan was frequency- and phase-aligned by spectral registration (Near et al., Reference Near, Edden, Evans, Paquin, Harris and Jezzard2015) using the FID-A toolkit (https://github.com/cic-methods/fid-a) in MATLAB (MathWorks, Natick, MA, USA) to minimize the effects of frequency and phase drifts. Motion-corrupted averages were removed as described previously (Near et al., Reference Near2013; Simpson et al., Reference Simpson, Devenyi, Jezzard, Hennessy and Near2017). Averaged difference spectra and sum spectra were then line-broadened using a 5-Hz Lorentzian filter, and zero-order phase corrections were applied manually to ensure upright peaks. First-order phase corrections were also applied manually in some cases because the first point in the FID did not always correspond exactly to the top of the echo. Following preprocessing, GABA and Glx signals from the difference spectra and creatine (Cr) signals from the sum spectra were quantified using the AMARES package provided in jMRUI software (Naressi et al., Reference Naressi, Couturier, Castang, de Beer and Graveron-Demilly2001; Vanhamme et al., Reference Vanhamme, Sundin, Hecke and Huffel2001). GABA, Glx, and Cr were modeled as a triplet, a doublet, and a singlet of Lorentzian peaks, respectively. Because the editing efficacy for Glx has not been determined in the current sequence, Glx signals are shown in an arbitrary unit value.

Measurement of MD Within MRS VOIs

Diffusion tensor data were converted into the NIfTI format using MRI Convert (http://lcni.uoregon.edu/~jolinda/MRIConvert). The data were preprocessed using FMRIB Software Library (FSL) version 4.1.5 (http://www.fmrib.ox.ac.uk/fsl). This procedure included the following: (1) eddy current correction, (2) motion correction by registering all the diffusion-weighted data to the b = 0 images, which were corrected first, (3) brain extraction, (4) calculation of diffusion tensor and diagonalization, and (5) transformation to the MNI space. After these processes, the MD map in the MNI space was automatically constructed using FSL.

Then, the obtained T1-weighted anatomical images were segmented into GM, WM, and cerebrospinal fluid (CSF) using FAST (FMRIB’s automated segmentation tool) (Zhang et al., Reference Zhang, Brady and Smith2001) in FSL software (available from http://fsl.fmrib.ox.ac.uk.proxy.bib.uottawa.ca/fsl/fslwiki/) (Jenkinson et al., Reference Jenkinson, Beckmann, Behrens, Woolrich and Smith2012; Smith et al., Reference Smith2004; Woolrich et al., Reference Woolrich2009; Zhang et al., Reference Zhang, Brady and Smith2001) to calculate the relative volume of each tissue within MRS VOIs. After the binary GM masks within MRS VOIs were formed from segmented GM images and MRS VOIs (Fig. 2), Statistical Parametric Mapping (SPM; http://www.fil.ion.ucl.ac.uk/spm/) was used to coregister different spaces between the binary GM masks and the MD map in each subject. The threshold of GM mask was set to the default SPM parameter of 0.8. Because MD is markedly higher in CSF than in the brain tissue, the map of MDs lower than, 0.102 × 10−2 mm2/s was generated to attenuate the CSF effect on MD measurement within MRS VOIs, in accordance with the previous method described by Albrecht et al. (Reference Albrecht2007), i.e. the above-mentioned cut-off values were set at 3 SDs above the mean measured tissue MD. Finally, we calculated mean MD from nonzero voxels of the MD map within binary GM masks of MRS VOIs, i.e. mPFC and PCC.

Figure 2. Binary mask of GM within mPFC VOI for measurement of MD. The figure shows a segmented GM image together with mPFC VOI in MRS (yellow) and a binary mask of GM within mPFC VOI in MRS (red) in the original space.

Statistical analysis

The metabolite ratios are presented in the results section of this study, i.e. raw GABA/Cr and Glx/Cr, and GM corrected GABA/Cr and Glx/Cr, which were divided by the relative volume of GM within each VOI in MRS in accordance with previous studies (Fujihara et al., Reference Fujihara2015; Jocham et al., Reference Jocham, Hunt, Near and Behrens2012). To assess the relationship among demographic characteristics, CANTAB scores, and metabolite ratios in MRS, Pearson’s correlation test were performed. In addition, we also performed regression analysis, in accordance with a previous report (Yildiz et al., Reference Yildiz2014). A p-value of < 0.05 was set as statistically significant. Student’s t-test was conducted to estimate gender differences in demographic characteristics. All the statistical analyses were carried out using SPSS version-24 (IBM Corporation).

Results

All the subjects showed MoCA scores ≥ 20, and the predicted IQs were higher than 90 in all the participants (mean ± SD = 112.2 ± 7.6; see Table 1). Pearson’s correlation test showed that in mPFC and PCC, NAA level, Cr level, raw GABA/Cr or Glx, or GM corrected GABA/Cr or Glx/Cr did not significantly correlate with age, education years, predicted IQ, MoCA score, or CANTAB test scores, i.e. SRM and RVIP scores. Furthermore, MD in mPFC and PCC did not significantly correlate with age, education years, predicted IQ, MoCA score, or the above-mentioned CANTAB test scores. The MD in mPFC or PCC did not correlate with Cr level in each VOI, respectively.

The raw Glx/Cr levels in mPFC and PCC were significantly negatively correlated with MD in each VOI (r = −0.477, p = 0.008 and r = −0.486, p = 0.004, respectively). The GM corrected Glx/Cr levels in mPFC and PCC were also significantly negatively correlated with MD in each VOI (r = −0.443, p = 0.014 and r = −0.487, p = 0.004, respectively) (Fig. 3). No significant correlations were observed between MD in mPFC and Glx level in PCC and between MD in PCC and Glx level in mPFC. Furthermore, raw or GM corrected GABA/Cr in mPFC and PCC did not significantly correlate with MD in either VOI.

Figure 3. Scatter plots showing correlation between cortical MD and GM corrected Glx/Cr in mPFC and PCC. Squares indicate male subjects and circles indicate female subjects.

In regression analysis, the raw Glx/Cr in mPFC and PCC could predict MD in each VOI (β = −0.477, t = −2.871, p = 0.008, and β = −0.486, t = −3.94, p = 0.004). Also, the GM corrected Glx/Cr in mPFC and PCC could predict MD in each VOI (β = −0.443, t = −2.617, p = 0.014, and β = −0.487, t = −3.102, p = 0.004). To control further for possible effects of GM, WM and CSF fractions were used in separate regression models for raw Glx/Cr, as reported by Fujihara et al. (Reference Fujihara2015) and Yildiz et al. (Reference Yildiz2014). All these models revealed that neither GM, WM, nor CSF explained the further variance in the models (mPFC: |β| < 0.463, |t| < 1.064, p > 0.297; and PCC: |β| < 0.131, |t| < 0.617, p > 0.542). These regression models showed that only raw Glx/Cr has a predictive effect on MD in each VOI (mPFC: β = −0.471, t = −2.724, p = 0.011; and PCC: β = −0.478, t = −2.908, p = 0.007). Also, GM corrected Glx/Cr has a predictive effect on MD in each VOI (mPFC: β = −0.422, t = −2.68, p = 0.021; and PCC: β = −0.477, t = −2.975, p = 0.006).

Discussion

The results of this study show that cortical MD is significantly negatively correlated with Glx levels in both mPFC and PCC in healthy elderly individuals. On the other hand, no significant association was found between GABA level and cortical MD in both VOIs.

Glx level is very likely related to excitatory neurotransmission, because glutamate is considered to be the major component of the Glx signal (Bauer et al., Reference Bauer2013; Fujihara et al., Reference Fujihara2015). The dominant pathway for glutamine production is via metabolism of neurotransmitter glutamate, so Glx represents the integrated metabolic and neurotransmitter functions of glutamate in the brain (Jahng et al., Reference Jahng2016; Yuksel and Ongur, Reference Yuksel and Ongur2010). The glutamate level in a certain GM region is suggested to be associated with neuronal dysfunction and loss in that GM region (Segovia et al., Reference Segovia, Porras, Del Arco and Mora2001). Thus, previous MRS studies showed that a decreased Glx level can be observed in neurodegenerative conditions, such as aging (Goryawala et al., Reference Goryawala, Sheriff and Maudsley2016; Grachev et al., Reference Grachev, Swarnkar, Szeverenyi, Ramachandran and Apkarian2001), mild cognitive impairment and Alzheimer’s disease (Antuono et al., Reference Antuono, Jones, Wang and Li2001; Huang et al., Reference Huang, Liu, Yin, Qian, Shrestha and Ni2016; Riese et al., Reference Riese2015), in several GM regions including mPFC and PCC, which implies neuronal dysfunction and loss in these GM regions. Consistently, cognitive performance or cognitive symptoms have been reported to positively correlate with Glx levels in elderly people (Zahr et al., Reference Zahr, Mayer, Pfefferbaum and Sullivan2008) and patients with mild cognitive impairment (Nikolova et al., Reference Nikolova, Stark and Stark2017) and Alzheimer’s disease (Walecki et al., Reference Walecki, Barcikowska, Cwikla and Gabryelewicz2011). Then, an increased cortical MD, which has been observed in aging and the above-mentioned diseases (Lin et al., Reference Lin2016; Ni et al., Reference Ni, Chen, Liu, Huang, Shen and Chen2010; Ray et al., Reference Ray2006; Weston et al., Reference Weston, Simpson, Ryan, Ourselin and Fox2015), may reflect the decline in the magnitude of microstructural organization in the GM associated with aging and age-related diseases, i.e. a breakdown of cytoarchitectural barriers such as the cell membrane or a shift in the concentration of water between intra- and extracellular spaces (Elman et al., Reference Elman2017; Neil et al., Reference Neil, Miller, Mukherjee and Huppi2002; Sundgren et al., Reference Sundgren, Dong, Gomez-Hassan, Mukherji, Maly and Welsh2004; Van Camp et al., Reference Van Camp2012). Considering the above-mentioned previous reports on Glx level and cortical MD, the significant negative association between cortical MD and Glx level in mPFC and PCC observed in this study suggests that age-related neuronal loss, which can be detected on the basis of cortical MD in DTI, leads to the decline of Glx metabolism within the above-mentioned GM regions.

In this study, no significant association was found between GABA levels and cortical MD in both mPFC and PCC. Similarly to Glx levels, previous MRS studies have shown that the GABA levels in mPFC and PCC are lower in healthy elderly people and patients with Alzheimer’s disease and mild cognitive impairments than in controls (Bai et al., Reference Bai2015; Riese et al., Reference Riese2015). Moreover, cognitive dysfunction has been reported to be associated with decreased GABA levels in mild cognitive impairment and Alzheimer’s disease (Porges et al., Reference Porges2017). Although such observations suggest that not only Glx levels, but also GABA levels reflect age-related neuronal loss, this speculation is inconsistent with the findings of our study. There are some difficulties in explaining the different findings on Glx and GABA levels in this study, but an asymmetric degeneration of glutamatergic neurons and GABAergic neurons during aging might contribute to our results regarding GABA in this study, that is, terminals and synapses of glutamatergic neurons, not those of GABAergic neurons, are reported to be predominantly affected in early-stage Alzheimer’s disease (Huang et al., Reference Huang, Liu, Yin, Qian, Shrestha and Ni2016; Kashani et al., Reference Kashani, Lepicard, Poirel, Videau, David, Fallet-Bianco, Simon, Delacourte, Giros, Epelbaum, Betancur and El Mestikawy2008; Proctor et al., Reference Proctor, Coulson and Dodd2010). Additional studies including age-related neurodegenerative disorders, i.e. mild cognitive impairment and Alzheimer’s disease, should be carried out. Furthermore, the lack of significant GABA observations might have been due to the small sample size in this study.

A major limitation in this study is its small sample size, which may have influenced our statistical results. In addition, owing to the lack of data from subjects with age-related neurodegenerative disorders, i.e. mild cognitive impairment or Alzheimer’s disease, there are some limitations in this study. Although some previous MRS studies have shown the significant associations between cognitive function test scores and the levels of several cortical metabolites, i.e. NAA, GABA or Glx (Jessen et al., Reference Jessen2013; Porges et al., Reference Porges2017), we failed to find such significant associations, which may be due to the small number of subjects in this study. In addition, young subjects should also be enrolled. Further MRS studies with larger sample sizes including subjects of various ages and with age-related neurodegenerative disorders will reinforce our findings in the future.

In conclusion, the level of Glx, which reflects excitatory neurotransmission, was associated with cortical MD in mPFC and PCC in GM. These findings might support a hypothesis, i.e. neuronal loss detected on the basis of cortical MD in DTI can cause a decline of Glx metabolism in these GM regions in elderly people. The results of this study can lead to a better understanding of the association between microstructural organization and neuronal metabolism in GM during aging.

Conflict of interest

None.

Description of authors’ roles

T. Motegi collected and analyzing the data and wrote the paper. K. Narita was responsible for the statistical design of the study and for carrying out the statistical analysis. K. Fujihara, M. Kasagi, Y. Suzuki and M. Tagawa collected the data and assisted with analyzing the data. K. Ujita helped carrying the study out. J. Near contributed to MRI and MRS methods and assisted with writing the article. M. Fukuda supervised the data collection and assisted with writing the article.

Acknowledgment

This work was supported by JSPS KAKENHI JP Grant Numbers 23591695 and 15KK0295.

References

Abe, O. et al. (2008). Aging in the CNS: comparison of gray/white matter volume and diffusion tensor data. Neurobiology of Aging, 29, 102116.CrossRefGoogle ScholarPubMed
Albrecht, J. et al. (2007). Voxel based analyses of diffusion tensor imaging in Fabry disease. Journal of Neurology, Neurosurgery, and Psychiatry, 78, 964969.CrossRefGoogle ScholarPubMed
Antuono, P.G., Jones, J.L., Wang, Y. and Li, S.J., (2001). Decreased glutamate + glutamine in Alzheimer’s disease detected in vivo with (1)H-MRS at 0.5 T. Neurology, 56, 737742.CrossRefGoogle ScholarPubMed
Bai, X. et al. (2015). Decreased gamma-aminobutyric acid levels in the parietal region of patients with Alzheimer’s disease. Journal of Magnetic Resonance Imaging, 41, 13261331.CrossRefGoogle ScholarPubMed
Bauer, J. et al. (2013). Craving in alcohol-dependent patients after detoxification is related to glutamatergic dysfunction in the nucleus accumbens and the anterior cingulate cortex. Neuropsychopharmacology, 38, 14011408.CrossRefGoogle ScholarPubMed
Benedetti, B. et al. (2006). Influence of aging on brain gray and white matter changes assessed by conventional, MT, and DT MRI. Neurology, 66, 535539.CrossRefGoogle ScholarPubMed
de Kouchkovsky, I., Fieremans, E., Fleysher, L., Herbert, J., Grossman, R.I. and Inglese, M., (2016). Quantification of normal-appearing white matter tract integrity in multiple sclerosis: a diffusion kurtosis imaging study. Journal of Neurology, 263, 11461155.CrossRefGoogle ScholarPubMed
Elman, J.A. et al. (2017). Genetic and environmental influences on cortical mean diffusivity. Neuroimage, 146, 9099.CrossRefGoogle ScholarPubMed
Fujihara, K. et al. (2015). Relationship of gamma-aminobutyric acid and glutamate+glutamine concentrations in the perigenual anterior cingulate cortex with performance of Cambridge Gambling Task. Neuroimage, 109, 102108.CrossRefGoogle ScholarPubMed
Garcia-Lazaro, H.G., Becerra-Laparra, I., Cortez-Conradis, D. and Roldan-Valadez, E., (2016). Global fractional anisotropy and mean diffusivity together with segmented brain volumes assemble a predictive discriminant model for young and elderly healthy brains: a pilot study at 3T. Functional Neurology, 31, 3946.Google Scholar
Goryawala, M.Z., Sheriff, S. and Maudsley, A.A., (2016). Regional distributions of brain glutamate and glutamine in normal subjects. NMR in Biomedicine, 29, 11081116.CrossRefGoogle ScholarPubMed
Grachev, I.D., Swarnkar, A., Szeverenyi, N.M., Ramachandran, T.S. and Apkarian, A.V., (2001). Aging alters the multichemical networking profile of the human brain: an in vivo (1)H-MRS study of young versus middle-aged subjects. Journal of Neurochemistry, 77, 292303.CrossRefGoogle Scholar
Huang, D., Liu, D., Yin, J., Qian, T., Shrestha, S. and Ni, H., (2016). Glutamate-glutamine and GABA in brain of normal aged and patients with cognitive impairment. European Radiology, 27, 26982705. CrossRefGoogle ScholarPubMed
Ihara, M., Okamoto, Y. and Takahashi, R., (2013). Suitability of the Montreal cognitive assessment versus the mini-mental state examination in detecting vascular cognitive impairment. Journal of Stroke and Cerebrovascular Diseases, 22, 737741.CrossRefGoogle ScholarPubMed
Jahng, G.H. et al. (2016). Glutamine and glutamate complex, as measured by functional magnetic resonance spectroscopy, alters during face-name association task in patients with mild cognitive impairment and Alzheimer’s disease. Journal of Alzheimer’s Disease, 52, 145159.CrossRefGoogle ScholarPubMed
Jenkinson, M., Beckmann, C.F., Behrens, T.E., Woolrich, M.W. and Smith, S.M., (2012). FSL. Neuroimage, 62, 782790.CrossRefGoogle ScholarPubMed
Jessen, F. et al. (2013). N-acetylaspartylglutamate (NAAG) and N-acetylaspartate (NAA) in patients with schizophrenia. Schizophrenia Bulletin, 39, 197205.CrossRefGoogle Scholar
Jiang, J. et al. (2017). Microstructural brain abnormalities in medication-free patients with major depressive disorder: a systematic review and meta-analysis of diffusion tensor imaging. Journal of Psychiatry & Neuroscience, 42, 150163.CrossRefGoogle ScholarPubMed
Jocham, G., Hunt, L.T., Near, J. and Behrens, T.E., (2012). A mechanism for value-guided choice based on the excitation-inhibition balance in prefrontal cortex. Nature Neuroscience, 15, 960961.CrossRefGoogle ScholarPubMed
Kashani, A., Lepicard, E., Poirel, O., Videau, C., David, J.P., Fallet-Bianco, C., Simon, A., Delacourte, A., Giros, B., Epelbaum, J., Betancur, C. and El Mestikawy, S., (2008). Loss of VGLUT1 and VGLUT2 in the prefrontal cortex is correlated with cognitive decline in Alzheimer disease. Neurobiology of Aging, 29, 16191630.CrossRefGoogle ScholarPubMed
Le Bihan, D., Breton, E., Lallemand, D., Grenier, P., Cabanis, E. and Laval-Jeantet, M., (1986). MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology, 161, 401407.CrossRefGoogle ScholarPubMed
Le Bihan, D. et al. (2001). Diffusion tensor imaging: concepts and applications. Journal of Magnetic Resonance Imaging, 13, 534546.CrossRefGoogle ScholarPubMed
Lin, S.H. et al. (2016). Increased Water Diffusion in the Parcellated Cortical Regions from the Patients with Amnestic Mild Cognitive Impairment and Alzheimer’s Disease. Frontiers in Aging Neuroscience, 8, 325.Google ScholarPubMed
Manna, A., Piras, F., Caltagirone, C., Bossu, P., Sensi, S.L. and Spalletta, G., (2015). Left hippocampus-amygdala complex macro- and microstructural variation is associated with BDNF plasma levels in healthy elderly individuals. Brain and Behavior, 5, e00334.CrossRefGoogle ScholarPubMed
Matsuoka, K., Uno, M., Kasai, K., Koyama, K. and Kim, Y., (2006). 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 and Clinical Neurosciences, 60, 332339.CrossRefGoogle ScholarPubMed
Mescher, M., Merkle, H., Kirsch, J., Garwood, M. and Gruetter, R., (1998). Simultaneous in vivo spectral editing and water suppression. NMR in Biomedicine, 11, 266272.3.0.CO;2-J>CrossRefGoogle ScholarPubMed
Morris, R.G., Evenden, J.L., Sahakian, B.J. and Robbins, T.W., (1987). Computer-aided assessment of dementia: comparative studies of neuropsychological deficits in Alzheimer-type dementia and Parkinson’s disease. In: Stahl, S., Iversen, S., Goodman, E. (Eds.), Cognitive Neurochemistry (pp. 2136). Oxford: Oxford University Press.Google Scholar
Naressi, A., Couturier, C., Castang, I., de Beer, R. and Graveron-Demilly, D., (2001). Java-based graphical user interface for MRUI, a software package for quantitation of in vivo/medical magnetic resonance spectroscopy signals. Computers in Biology and Medicine, 31, 269286.CrossRefGoogle ScholarPubMed
Nasreddine, Z.S. et al. (2005). The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53, 695699.CrossRefGoogle ScholarPubMed
Near, J. et al. (2013). Unedited in vivo detection and quantification of gamma-aminobutyric acid in the occipital cortex using short-TE MRS at 3 T. NMR in Biomedicine, 26, 13531362.CrossRefGoogle ScholarPubMed
Near, J., Edden, R., Evans, C.J., Paquin, R., Harris, A. and Jezzard, P., (2015). Frequency and phase drift correction of magnetic resonance spectroscopy data by spectral registration in the time domain. Magnetic Resonance in Medicine, 73, 4450.CrossRefGoogle ScholarPubMed
Neil, J., Miller, J., Mukherjee, P. and Huppi, P.S., (2002). Diffusion tensor imaging of normal and injured developing human brain - a technical review. NMR in Biomedicine, 15, 543552.CrossRefGoogle Scholar
Nelson, H.E., (1982). National Adult Reading Test (NART): For the assessment of premorbid intelligence in patients with dementia: Test manual. NFER-Nelson. Windsor, United Kingdom.Google Scholar
Nesteruk, T., Nesteruk, M., Styczynska, M., Barcikowska-Kotowicz, M. and Walecki, J., (2016). Radiological evaluation of strategic structures in patients with mild cognitive impairment and early Alzheimer’s disease. Polish Journal of Radiology, 81, 288294.CrossRefGoogle ScholarPubMed
Ni, J.M., Chen, S., Liu, J.J., Huang, G., Shen, T.Z. and Chen, X.R., (2010). Regional diffusion changes of cerebral grey matter during normal aging--a fluid-inversion prepared diffusion imaging study. European Journal of Radiology, 75, 134138.CrossRefGoogle ScholarPubMed
Nikolova, S., Stark, S.M. and Stark, C.E.L., (2017). 3T hippocampal glutamate-glutamine complex reflects verbal memory decline in aging. Neurobiology of Aging, 54, 103111.CrossRefGoogle Scholar
Nishioka, C., Poh, C. and Sun, S.W., (2015). Diffusion tensor imaging reveals visual pathway damage in patients with mild cognitive impairment and Alzheimer’s disease. Journal of Alzheimer’s Disease, 45, 97107.CrossRefGoogle ScholarPubMed
Oldfield, R.C., (1971). The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 9, 97113.CrossRefGoogle ScholarPubMed
Porges, E.C. et al. (2017). Frontal gamma-aminobutyric acid concentrations are associated with cognitive performance in older adults. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2, 3844.Google ScholarPubMed
Proctor, D.T., Coulson, E.J. and Dodd, P.R., (2010). Reduction in post-synaptic scaffolding PSD-95 and SAP-102 protein levels in the Alzheimer inferior temporal cortex is correlated with disease pathology. Journal of Alzheimer’s Disease, 21, 795811.CrossRefGoogle ScholarPubMed
Ray, K.M. et al. (2006). Mild cognitive impairment: apparent diffusion coefficient in regional gray matter and white matter structures. Radiology, 241, 197205.CrossRefGoogle ScholarPubMed
Riese, F. et al. (2015). Posterior cingulate gamma-aminobutyric acid and glutamate/glutamine are reduced in amnestic mild cognitive impairment and are unrelated to amyloid deposition and apolipoprotein E genotype. Neurobiology of Aging, 36, 5359.CrossRefGoogle ScholarPubMed
Rose, S.E., Janke, A.L. and Chalk, J.B., (2008). Gray and white matter changes in Alzheimer’s disease: a diffusion tensor imaging study. Journal of Magnetic Resonance Imaging, 27, 2026.CrossRefGoogle ScholarPubMed
Segovia, G., Porras, A., Del Arco, A. and Mora, F., (2001). Glutamatergic neurotransmission in aging: a critical perspective. Mechanisms of Ageing and Development, 122, 129.CrossRefGoogle ScholarPubMed
Simpson, R., Devenyi, G.A., Jezzard, P., Hennessy, T.J. and Near, J., (2017). Advanced processing and simulation of MRS data using the FID-Appliance (FID-A) – An open source, MATLAB based toolkit. Magnetic Resonance in Medicine, 77, 2333.CrossRefGoogle ScholarPubMed
Singh, S. et al. (2016). Microstructural abnormalities of uncinate fasciculus as a function of impaired cognition in schizophrenia: A DTI study. Journal of Biosciences, 41, 419426.CrossRefGoogle ScholarPubMed
Smith, S.M. et al. (2004). Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage, 23, Suppl 1, S208219.CrossRefGoogle ScholarPubMed
Sundgren, P.C., Dong, Q., Gomez-Hassan, D., Mukherji, S.K., Maly, P. and Welsh, R., (2004). Diffusion tensor imaging of the brain: review of clinical applications. Neuroradiology, 46, 339350.CrossRefGoogle ScholarPubMed
Van Camp, N. et al. (2012). A complementary diffusion tensor imaging (DTI)-histological study in a model of Huntington’s disease. Neurobiology of Aging, 33, 945959.CrossRefGoogle Scholar
Vanhamme, L., Sundin, T., Hecke, P.V. and Huffel, S.V., (2001). MR spectroscopy quantitation: a review of time-domain methods. NMR in Biomedicine, 14, 233246.CrossRefGoogle ScholarPubMed
Waldron-Perrine, B. and Axelrod, B.N., (2012). Determining an appropriate cutting score for indication of impairment on the Montreal Cognitive Assessment. International Journal of Geriatric Psychiatry, 27, 11891194.CrossRefGoogle ScholarPubMed
Walecki, J., Barcikowska, M., Cwikla, J.B. and Gabryelewicz, T., (2011). N-acetylaspartate, choline, myoinositol, glutamine and glutamate (glx) concentration changes in proton MR spectroscopy (1H MRS) in patients with mild cognitive impairment (MCI). Medical Science Monitor, 17, Mt105111.CrossRefGoogle Scholar
Weston, P.S., Simpson, I.J., Ryan, N.S., Ourselin, S. and Fox, N.C., (2015). Diffusion imaging changes in grey matter in Alzheimer’s disease: a potential marker of early neurodegeneration. Alzheimer’s Research & Therapy, 7, 47.CrossRefGoogle ScholarPubMed
Woolrich, M.W. et al. (2009). Bayesian analysis of neuroimaging data in FSL. Neuroimage, 45, S173186.CrossRefGoogle ScholarPubMed
Yildiz, A. et al. (2014). Feeling safe in the plane: neural mechanisms underlying superior action control in airplane pilot trainees—a combined EEG/MRS study. Human Brain Mapping, 35, 50405051.CrossRefGoogle Scholar
Yuksel, C. and Ongur, D., (2010). Magnetic resonance spectroscopy studies of glutamate-related abnormalities in mood disorders. Biological Psychiatry, 68, 785794.CrossRefGoogle ScholarPubMed
Zahr, N.M., Mayer, D., Pfefferbaum, A. and Sullivan, E.V., (2008). Low striatal glutamate levels underlie cognitive decline in the elderly: evidence from in vivo molecular spectroscopy. Cerebral Cortex, 18, 22412250.CrossRefGoogle ScholarPubMed
Zhang, Y., Brady, M, and Smith, S., (2001). Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Transactions on Medical Imaging, 20, 4557.CrossRefGoogle ScholarPubMed
Ziyan, U. and Westin, C.F., (2008). Joint segmentation of thalamic nuclei from a population of diffusion tensor MR images. Medical Image Computing and Computer-Assisted Intervention, 11, 279286.Google ScholarPubMed
Figure 0

Table 1. Demographic characteristics of study subjects

Figure 1

Figure 1. Magnetic resonance spectroscopy (MRS) using MEGA-PRESS. The volumes of interest (30 × 20 × 20 mm3) in MRS, which were located on mPFC and PCC, are shown in A; “line a” is set exactly on the rostral margin of the corpus callosum as the perpendicular axis to the anterior commissure–posterior commissure (AC–PC) line. “line b” is drawn as the perpendicular axis through the anterior commissure to the AC–PC line. The edited spectrum (i.e. black line) and fitted curve (i.e. red line), which were obtained for the signal quantification of GABA, Glx, and Cr levels, are shown in B.

Figure 2

Figure 2. Binary mask of GM within mPFC VOI for measurement of MD. The figure shows a segmented GM image together with mPFC VOI in MRS (yellow) and a binary mask of GM within mPFC VOI in MRS (red) in the original space.

Figure 3

Figure 3. Scatter plots showing correlation between cortical MD and GM corrected Glx/Cr in mPFC and PCC. Squares indicate male subjects and circles indicate female subjects.