Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Ultra high-field (7tesla) magnetic resonance spectroscopy in Amyotrophic Lateral Sclerosis

  • Nazem Atassi,

    Affiliation Neurological Clinical Research Institute (NCRI), Department of Neurology, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, United States of America

  • Maosheng Xu,

    Affiliation Department of Radiology, First Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou, P.R. China

  • Christina Triantafyllou,

    Current address: Siemens Healthineers, Erlangen, Germany

    Affiliation Department of Radiology, Massachusetts General Hospital, Athinoula A. Martinos Center for Biomedical Imaging, and Harvard Medical School, Charlestown, Massachusetts, United States of America

  • Boris Keil,

    Current address: Institute for Medical Physics and Radiation Safety, Dept. of Life Science Engineering, Mittelhessen University of Applied Sciences, Gießen, Germany

    Affiliation Department of Radiology, Massachusetts General Hospital, Athinoula A. Martinos Center for Biomedical Imaging, and Harvard Medical School, Charlestown, Massachusetts, United States of America

  • Robert Lawson,

    Affiliation Neurological Clinical Research Institute (NCRI), Department of Neurology, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, United States of America

  • Paul Cernasov,

    Affiliation Neurological Clinical Research Institute (NCRI), Department of Neurology, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, United States of America

  • Elena Ratti,

    Current address: Biogen, Cambridge, Massachusetts, United States of America

    Affiliation Neurological Clinical Research Institute (NCRI), Department of Neurology, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, United States of America

  • Christopher J. Long,

    Affiliation Massachusetts Institute of Technology, Sloan School of Management, Cambridge, Massachusetts, United States of America

  • Sabrina Paganoni,

    Affiliation Department of Physical Medicine and Rehabilitation, Spaulding Rehabilitation Hospital, Charlestown, Massachusetts, United States of America

  • Alyssa Murphy,

    Affiliation Neurological Clinical Research Institute (NCRI), Department of Neurology, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, United States of America

  • Nouha Salibi,

    Affiliation Siemens Healthineers, MR R&D, Siemens, Auburn, Alabama, United States of America

  • Ravi Seethamraju,

    Affiliation Siemens Healthineers, MR R&D, Siemens, Charlestown, Massachusetts, United States of America

  • Bruce Rosen,

    Affiliation Department of Radiology, Massachusetts General Hospital, Athinoula A. Martinos Center for Biomedical Imaging, and Harvard Medical School, Charlestown, Massachusetts, United States of America

  • Eva-Maria Ratai

    ratai@nmr.mgh.harvard.edu

    Affiliation Department of Radiology, Massachusetts General Hospital, Athinoula A. Martinos Center for Biomedical Imaging, and Harvard Medical School, Charlestown, Massachusetts, United States of America

Ultra high-field (7tesla) magnetic resonance spectroscopy in Amyotrophic Lateral Sclerosis

  • Nazem Atassi, 
  • Maosheng Xu, 
  • Christina Triantafyllou, 
  • Boris Keil, 
  • Robert Lawson, 
  • Paul Cernasov, 
  • Elena Ratti, 
  • Christopher J. Long, 
  • Sabrina Paganoni, 
  • Alyssa Murphy
PLOS
x

Abstract

The main objective of this study was to utilize high field (7T) in vivo proton magnetic resonance imaging to increase the ability to detect metabolite changes in people with ALS, specifically, to quantify levels of glutamine and glutamine separately. The second objective of this study was to correlate metabolic markers with clinical outcomes of disease progression. 13 ALS participants and 12 age-matched healthy controls (HC) underwent 7 Tesla MRI and MRS. Single voxel MR spectra were acquired from the left precentral gyrus using a very short echo time (TE = 5 ms) STEAM sequence. MRS data was quantified using LCModel and correlated to clinical outcome markers. N-acetylaspartate (NAA) and total NAA (tNA, NAA + NAAG) were decreased by 17% in people with ALS compared to HC (P = 0.004 and P = 0.005, respectively) indicating neuronal injury and/or loss in the precentral gyrus. tNA correlated with disease progression as measured by forced vital capacity (FVC) (P = 0.014; Rρ = 0.66) and tNA/tCr correlated with overall functional decline as measured by worsening of the ALS Functional Rating Scale-Revised (ALSFRS-R) (P = 0.004; Rρ = -0.74). These findings underscore the importance of NAA as a reliable biomarker for neuronal injury and disease progression in ALS. Glutamate (Glu) was 15% decreased in people with ALS compared to HC (P = 0.02) while glutamine (Gln) concentrations were similar between the two groups. Furthermore, the decrease in Glu correlated with the decrease in FVC (P = 0.013; Rρ = 0.66), a clinical marker of disease progression. The decrease in Glu is most likely driven by intracellular Glu loss due to neuronal loss and degeneration. Neither choline containing components (Cho), a marker for cell membrane turnover, nor myo-Inositol (mI), a suspected marker for neuroinflammation, showed significant differences between the two groups. However, mI/tNA was correlated with upper motor neuron burden (P = 0.004, Rρ = 0.74), which may reflect a relative increase of activated microglia around motor neurons. In summary, 7T 1H MRS is a powerful non-invasive imaging technique to study molecular changes related to neuronal injury and/or loss in people with ALS.

Introduction

Amyotrophic Lateral Sclerosis (ALS) is a progressive neurodegenerative disorder affecting the upper and lower motor neurons leading to muscle atrophy, weakness, and death due to respiratory failure. To this date, no treatment prevents, halts, or reverses the disease; though the FDA-approved drug riluzole provides a modest survival benefit. The pathogenesis of ALS is not very well understood [1]. Proposed pathogenic mechanisms include mitochondrial dysfunction, glutamate-mediated excitotoxicity, endoplasmic reticulum stress, free radical-mediated oxidative cytotoxicity, and neuroinflammation including microglia activation and astrogliosis [2]. Altogether, these mechanisms lead to progressive neuronal loss. Studies of the SOD1 rodent models of ALS suggest that glutamate excitotoxicity and neuroinflammation are the drivers of neuronal death and ALS progression [3, 4].

Proton magnetic resonance imaging (1H MRS) is a neuroimaging technique that allows quantifying metabolic changes in vivo. Thus, 1H MRS has the potential to provide surrogate markers to assess disease progression and monitor therapeutic treatments in ALS. Specifically, 1H MRS is able to quantify metabolites such as N-acetylaspartate (NAA), a marker of neuronal integrity; creatine (Cr), a marker of energy metabolism; choline-containing compounds (Cho), a marker of cell membrane turnover; myo-inositol (mI) a marker of glial activation, and several neurotransmitters and their precursors e.g. γ-aminobutyric acid (GABA), glutamate (Glu), and glutamine (Gln). Indeed, there is a vast body of literature relating to the evaluation of ALS using 1H MRS at 1.5 Tesla and 3 Tesla [59].

The majority studies have described a reduction in NAA concentrations [1012] or NAA ratios (e.g. NAA/Cr or NAA/Cho) [1321] in the motor cortex of people with ALS compared to heathy controls suggesting either neuronal loss or injury in the motor cortex of ALS patients. While measurements of NAA concentrations have provided to be reliable in ALS, measurements of other metabolites and neurotransmitters at 3T have been inconsistent.

Increased Cho and Cho/Cr levels have been reported in people with ALS [15, 2224] in some studies but not others [16, 21, 25, 26]. Elevated mI and mI/Cr have been detected in the motor cortex in people with ALS [15, 22, 27, 28] but not in early-stage ALS [29].

1H MRS reports on Glu and/or the sum of Glu and Gln are also variable. On MR scanners operating at 1.5T or 3.0T, the resonances of Glu and Gln overlap significantly and are commonly referred to as Glx. Early studies found no significant differences in Glx and Glu levels [10, 28] in the motor cortices, occipital cortex, and brain stem, while some other studies reported decreased Glu levels in the precentral gyrus in patients with ALS compared with controls [15]. Using a TE-averaged Point Resolved Spectroscopy (PRESS) sequence at 3T, Han et al. reported that both Glu/Cr and Glx/Cr were increased in the motor cortices as well as in the posterior limb of the internal capsule [30]. Furthermore, early magnetic resonance spectroscopic imaging (MRSI) studies by Pioro et. al reported increased levels Glx in the medulla in people with ALS compared to healthy controls [13, 14].

The objective of this study was to utilize ultra-high field (7T) 1H MRS to increase our ability to detect metabolite changes in people with ALS, specifically, to quantify levels of Glu and Gln separately. The second objective of this study was to correlate these metabolic markers with clinical outcomes of disease progression.

Materials and methods

The study was approved by the Partners Human Research Committee. All participants provided written informed consent.

Study participants

Fourteen people with ALS according to the revised El Escorial criteria and 12 age-matched healthy controls (HC) were enrolled and underwent 7T MRS. Eligibility criteria included being able to lie flat for at least 45 minutes and considered medically safe to tolerate MRI scanning. Healthy age-matched controls (HC) were recruited from our registry of healthy volunteers.

Mean age did not differ between HC (52±10 years old) and people with ALS (56±10 years old) (P = 0.21). Of the 26 participants, 13 participants with ALS (10 male, 3 female, 56±10 y/o, age range 37–76 years) and 12 healthy controls (6 male, 6 female, 52±10 y/o, age range 37–78 years) had good quality MRS data, and were included in the analysis. Baseline characteristics of the 13 ALS participants are summarized in Table 1.

Clinical measures included the worsening of the ALS Functional Rating Scale–revised (ALSFRS-R), forced vital capacity (FVC), and upper motor neuron burden (UMNB). The ALSFRS-R is a quickly administered (10 minutes) ordinal rating scale (ratings 0–4), which determines patients' assessment of their capability and independence in 12 functional activities including bulbar, gross motor scale (GMS), fine motor scale (FMS), and respiratory performance [31]. We estimated the rate of disease progression (ΔFS) using the following formula: (48—ALSFRS-R)/disease duration [32]. FVC is measured using a spirometry in an upright position, and is used to assess diaphragmatic weakness in people with ALS. The UMNB score reflects measurement of the following deep tendon (scores 0–4) and pathological reflexes (present—1 or absent—0): biceps, brachioradialis, triceps, knee jerk, ankle jerk, Hoffman, Babinski, and jaw jerk. The total UMNB score ranges from 0 to 45, with 0 representing areflexia and 45 representing the highest possible UMN burden.

Data acquisition

All in vivo MRI experiments were performed on a 7 Tesla MR imager (Siemens AG Erlangen) using a transmit birdcage coil and a 32-channel receive coil. MRS data was prescribed based on the axial, coronals, and sagittal reformatted images obtained by a multi-echo MPRAGE sequence with 1mm3 isotropic resolution, GRAPPA acceleration factor 2, TR = 2530 TE(1) = 1.44 ms, TE(2) = 3.18 ms, TE(3) = 5.04 ms, and TE(4) = 6.9 ms.

Single voxel 1H MR spectra were acquired from the motor cortex (voxel of interest (VOI) = 2x2x2cm3) using a very short echo time STEAM sequence with VAPOR water suppression technique (STEAM: STimulated Echo Acquisition Mode and VAPOR: Variable Power and Optimized Relaxation Delays) [33, 34]. Consistent voxel placement across subjects was confirmed via neuroanatomical identification of the precentral gyrus by the same neurologist for all scans (Fig 1).

thumbnail
Fig 1. Placement of a 2x2x2cm3 voxel on the left precentral gyrus overlaid on a sagittal, coronal and axial reformatted MPRAGE.

The voxel does not appear to be square as the voxel is oblique to the sagittal and axial view and the graph shows the voxels’ projections.

https://doi.org/10.1371/journal.pone.0177680.g001

Very short echo times of 5 ms were used to reduce T2 relaxation effects and J-modulation, which improves the quantification of Glu and Gln. High bandwidth slice-selective radiofrequency (RF) pulses were used to minimize chemical shift displacement errors. Other parameters included TR of 5000 ms, TM of 75 ms, bandwidth of 4k, pulse duration of 2.6 ms, and averages of 96. No outer volume saturation was used. Prior to data acquisition, the MRS ROI underwent an automatic shim routine based on gradient double acquisition (GRE-shim) using first and second order shims followed by first order manual shimming. Average full widths at half maximum (FWHM) of the NAA resonance was 11±1Hz. Water unsuppressed spectra from the same voxel were acquired to estimate absolute concentrations in institutional units (I.U.) and to compensate for eddy currents.

Spectral quantification

Metabolite quantification was performed with LCModel (version 6.3, Copyright S.W. Provencher) for the following 17 metabolites: alanine (Ala), aspartate (Asp), ascorbate (Asc), creatine (Cr), phosphocreatine (PCr), γ-aminobutyric acid (GABA), glutamine (Gln), glutamate (Glu), glutathione (GSH), glycerophosphocholine (GPC), phosphocholine (PCho), myo-inositol (mI), scyllo-inositol (scyllo-Ins), lactate (Lac), N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), and taurine (Tau). We also measured the ratios of tNA/tCr and mI/tNAA as those were reported as meaningful biomarkers in ALS. Specifically, mI/tNA provided an optimal sensitivity and specificity profile [27].

The 7T LC Model basis set was simulated by Dr. Dinesh Deelchand from the Center for Magnetic Resonance Research at the University of Minnesota using Matlab based on density matrix formalism and an updated database of chemical shifts and coupling constants [35, 36], as described previously [37]. Averaged spectra macromolecules using metabolite nulling inversion-recovery experiments (acquisition parameters: TR = 2s and TI = 675ms) were also included in the 7T LCModel basis sets. The LCModel analysis was performed on spectra within the chemical shift range 0–4ppm. The signal to noise ratio (SNR) was estimated in LCModel and is defined as the peak height of NAA divided by the root mean square of the noise of the LCModel fit. The control parameters that LC Model uses for calculation of metabolite concentration from a water reference were not changed from the default values. Two representative spectra of an ALS patient compared to an age-matched control are shown (Fig 2).

thumbnail
Fig 2. Magnetic resonance spectra representative spectra obtained with very short TE STEAM with VAPOR water suppression (TE 5 ms, TR 5000ms, TM 75 ms, 2 x 2 x 2 cm3 voxel size and 96 averages) of a healthy control (A) and an ALS patient (B).

Please note the excellent overlay between the acquired data and LCModel fit.

https://doi.org/10.1371/journal.pone.0177680.g002

The multi echo MPRAGE sequence was used to segment brain volume of white matter, gray matter, or cerebrospinal fluid (CSF) within the 2x2x2cm3 MRS voxel using FMRIB Software Library (FSL 5.0, Oxford, UK). All metabolite levels were adjusted for CSF contribution.

Statistical analysis

The primary focus of this study was 1) to determine the directionality of glutamate in the motor cortex of people with ALS and 2) to evaluate metabolic changes related to neuronal injury (NAA). For these primary hypnotizes, two-tailed Student t-tests were used to compare Glu and NAA concentrations between ALS subjects and healthy controls and multiple testing Bonferroni correction of P = 0.05/2 = 0.025 was made.

Other metabolites (NAAG, Gln, mI, GPC, GSH, tCr and GABA) were examined for further exploratory analyses. Again, two-tailed Student t-tests were used to compare metabolite concentrations between ALS subjects and healthy controls. For these additional exploratory metabolites, no corrections for multiple comparisons were made. All analyses were conducted using JMP 10.0 (SAS, Cary, NC). Spearman correlation coefficients were calculated to estimate the correlations between MRS markers and clinical outcomes, e.g. FVC, UMNB and worsening of the disease progression (ΔFS). For that, only metabolites/ metabolite ratios were selected for this analysis that showed a significant difference or a trend between people with ALS and HC. For the correlational analysis, multiple testing Bonferroni correction of P = 0.05/3 = 0.017 was made.

Results and discussion

Subjects

Fourteen people with ALS and 12 age-matched healthy controls (HC) were enrolled and underwent 7T MRS. All HC and ALS participants tolerated the ~45 min scan. However, ALS as well as HC participants reported dizziness when moved within the fringe field of the 7T, which subsided after lying still in the scanner.

VOI segmentation

Since the single voxel VOI encompassed CSF, gray matter, and white matter, the VOI of each subject was segmented using FSL. VOI included on average 61%±9% white matter, 28%±7% gray mater, and 11%±9% CSF. Of note, there was no significant difference between ALS and HC subjects in the ratios of CSF (P = 0.76), gray matter (P = 0.71) and white mater (P = 0.39). To verify that none of the compartments were either under- or overestimated, we overlaid the MRS voxel with the MNI template using FSL. While we have no gold standard to assess the performance of the FSL segmentation tool, we did manually inspect all of its outputs.

Metabolites

The SNR estimated in LCModel (defined as the peak height of NAA divided by the root mean square of the noise of the LCModel fit) was 40±7. Analyses are based on 13 ALS patients and 12 HC. One patient with ALS was excluded because he was unable to complete the MRS scan.

The comparison between ALS participants and healthy controls revealed a decrease of 17% in NAA in people with ALS compared to healthy controls (P = 0.004) (Table 2) while no changes in NAAG were observed (P = 0.25). Glutamate levels were 15% decreased in people with ALS (P = 0.02) while no changes in Gln were observed (P = 0.99). There was a correlation between NAA and Glu (P<0.0001, Rρ = 0.74, Fig 3), but not between NAA and Gln (P = 0.3, Rρ = 0.18). GABA, GSH, and tCr levels were not significantly different in ALS patients compared to the levels in HC, (P = 0.16, P = 0.14, and P = 0.16, respectively). Neither choline containing components nor mI levels showed significant differences between the two groups.

thumbnail
Table 2. Absolute metabolic concentrations (mM in institutional units) and relative metabolic concentrations of the left motor cortex in ALS patients vs. age-matched healthy controls.

https://doi.org/10.1371/journal.pone.0177680.t002

thumbnail
Fig 3. Correlation between NAA and glutamate (Glu) (P<0.0001, Rρ = 0.74) and between total NAA (tNA) and Glu (P<0.0001, Rρ = 0.88), blue circles representing heathy control and red circles representing ALS data.

https://doi.org/10.1371/journal.pone.0177680.g003

We also measured the ratios of tNA/tCr and mI/tNA as those were reported as meaningful biomarkers in ALS. Specifically, mI/tNA provided an optimal sensitivity and specificity profile [27]. tNA/Cr was 7% decreased in people with ALS, however, these differences only showed a trend toward significance (P = 0.06). Finally, people with ALS had 18% higher mI/tNA levels compared to healthy controls (P = 0.03).

In the appendix (S1 Table), we have extended our analysis to all 17 metabolites of the basis-set.

Correlations between metabolism and clinical measures

Lastly, we investigated if any of the clinical outcome measures in ALS correlated with the metabolic concentrations or metabolic ratios (Table 3). We only selected metabolites/ metabolite ratios that showed a significant difference or a trend between people with ALS and HC. FVC correlated negatively with tNA and Glu. Upper motor neuron burden (UMNB) correlated positively with mI/tNA (P = 0.004, Rρ = 0.74). Lastly, Spearman Rank correlations showed a negative relationship between ΔFS and tNA/tCr (P = 0.028, Rρ = -0.60).

thumbnail
Table 3. Correlations between clinical outcomes and metabolites/metabolite ratios in ALS.

https://doi.org/10.1371/journal.pone.0177680.t003

The development of in vivo biomarkers to aid in early diagnosis, to understand disease mechanisms, and to serve as readout of experimental treatments is an unmet need in ALS. Neuroimaging is an emerging candidate as a source of biomarkers for ALS. Here, we used 7 Tesla MRS to quantify brain metabolites and correlate them with ALS disease progression. Previous studies have demonstrated the ability of high field in vivo 1H MRS to increase detection sensitivity for those compared to lower field strengths [36]. In addition, we utilized an MRS sequence with very short echo times of 5 ms to reduce T2 relaxation effects and J-modulation, which improves the quantification of Glu and Gln.

Our findings of decreased NAA and tNA in ALS are consistent with the findings from previous 1.5T and 3T MRS studies in ALS [1012, 1421, 30, 38]. Within the adult brain, NAA is found predominantly in neurons, and serves as a marker of neuronal integrity and mitochondrial dysfunction [39]. Thus, the decrease in NAA can be attributed to either neuronal loss or injury. While most studies described a reduction in tNA concentrations or tNA ratios (tNA/tCr) in the motor cortex of people with ALS, a few recent reports showed that tNA relative to water did not significantly differ between people with ALS and HC, merely the ratio of NAA/Cr decreased [26, 40] suggesting small opposite changes in NAA and tCr in ALS. Our study reveals both tNA and tNA ratios over tCr are decreased in the motor cortex people with ALS. Furthermore, we could not confirm an increase in tCr in our study.

Spearman Rank correlations showed an inverse relationship between rate of disease progression (ΔFS) and tNA/tCr. These results indicate that neuronal injury/loss is more prominent in ALS participant in whom the disease is progressing at a faster rate. Furthermore, a decrease in tNA was associated with a decrease in forced vital capacity, another measure of disease progression. A number of studies have also explored correlations between tNA and clinical scales, and have reported significant associations among tNA levels and ALSFRS-R scores [12, 17, 41], UMNB burden scores [16, 17, 41], and rate of disease progression [42]. Thus, our study confirms the importance of NAA as reliable biomarker for neuronal injury and disease progression.

There is sufficient evidence to support that excitotoxicity may contribute to ALS pathophysiology [3, 4]. Glutamate is the major excitatory neurotransmitter in the CNS and is involved in the pathogenesis of several neurological disorders when its extracellular concentration rises to toxic levels, termed excitotoxicity [43, 44]. The amino acid Gln is a precursor and storage form of Glu, which is predominantly located in astrocytes [45]. Glutamate levels are increased in the serum and CSF of people with ALS [46]. Previous in vivo MRS studies in people with ALS used low magnetic fields (1.5 or 3 Tesla), and reported variable results on Glu and Glx [10, 15, 28, 30]. The development of higher magnetic field strengths makes this separation more feasible. Thus, ultra-high magnetic fields such as 7T are desired to accurately quantify glutamate and glutamine separately. Furthermore, we have utilized a STEAM sequence using very short echo times of 5 ms to reduce T2 relaxation effects and J-modulation, which improves the quantification of Glu and Gln.

Our study shows that glutamate levels are decreased in the ALS group compared to healthy controls, whereas glutamine levels are not different. Furthermore, Glu correlated positively with FVC, which is a measure of disease progression.

We hypothesize that the observed decrease in glutamate could be secondary to the loss of intracellular glutamate due to neuronal loss, as reflected in the decreased NAA levels. MRS cannot identify whether the source of metabolic concentrations is from intra- or extra-cellular spaces. Glutamate is predominantly an intracellular metabolite and its role in excitotoxicity probably stems from increases in its extracellular concentrations. However, Glu levels in CSF and in extracellular compartment are at least by three orders of magnitude lower than those in intracellular compartment [47], thus glutamate concentrations measured by MRS represent mostly the intracellular glutamate concentrations, and the reduction of glutamate we observed in the ALS group is probably secondary to intracellular glutamate depletion due to neuronal loss. This hypothesis is supported by the significant positive correlation between Glu and NAA. Glutamate predominantly shares the same compartment as NAA—namely neurons [4850]. Therefore, a strong correlation between Glu and NAA (Fig 3) could indicate neuronal loss. Our results are consistent with reported decrease in Glu or Glx in other neurodegenerative diseases including Parkinson’s disease [51] and Alzheimer’s disease [52].

In our study, none of the glial markers mI, GPC or tCho was significantly elevated in ALS patients compared to controls. However, we saw a positive correlation between the markers of glial proliferation/inflammation mI/NAA and clinical measures of upper motor neuron dysfunction (increased reflexes) in the ALS group. These findings are in line with ALS postmortem pathological studies showing activated microglia around motor neurons [5357]. Furthermore, a recent PET study revealed increased [11C]-PBR28 binding in the precentral gyrus in patients with ALS compared to controls reflecting increased glial activation [58].

This study has a few limitations that we like to discuss below:

  1. For the LCModel analysis. we did not change the default control parameters that LC Model uses for calculation of metabolite concentration from a water reference, which may explain why the reported metabolite concentrations are on the lower side. LC Model assumes a default concentration for water of 0.3588 assuming 100% white matter and a signal attenuation factor of 0.7 to account for T2 relaxation assuming a TE of ~ 30 ms. On the other hand, applying a different scaling factor to both groups will not change the statistics between people with ALS and heathy controls or the correlations between the metabolites and the clinical outcome markers.
  2. Of note, virtually all metabolite levels were decreased in ALS patients. To ensure, we simple did not simple measure atrophy in the motor cortex of ALS patients, we calculated CSF contribution of each voxel using FSL and corrected the metabolite concentrations for it. An alternative approach to eliminate CSF contributions as potential confounding factor is to report metabolic ratios, typically over total creatine. However, reporting ratios of creatine may be problematic as one cannot be certain that creatine is stable in ALS patients.
  3. One of the major limitations include the relatively small sample size and multiple testing which may result in type 1 error. The primary focus of our study was to evaluate metabolic changes related to neuronal injury (NAA) and to determine the directionality of Glu in the motor cortex of people with ALS; for these primary hypnotizes, multiple testing Bonferroni correction of P = 0.05/2 = 0.025 was made. Other metabolites were examined for further exploratory analyses.

For the correlation analyses, we limited tests between the clinical markers and MRS markers to only the metabolites/metabolite ratios that showed a significant difference or a trend between people with ALS and HC. Furthermore, multiple testing Bonferroni correction of P = 0.05/3 = 0.017 was made. We consider this work an important pilot and hypothesis-generating study that will require future confirmation and we plan to use strict correction for multiple testing in future larger study.

Conclusion

In conclusion, ultra-high field MRS is a powerful and non-invasive imaging technique to study molecular changes of neuronal injury in people with ALS. Future studies that include longitudinal imaging in people with ALS will allow to study metabolite changes as disease progresses, and will help clarify the role of ultra-high field MRS as a biomarker of target engagement for proof-of-mechanism clinical trials.

Supporting information

S1 Table. Absolute metabolic concentrations (mM in institutional units) of all 17 metabolites analyzed of the left motor cortex in ALS patients (N = 13) vs. age-matched healthy controls (N = 12).

https://doi.org/10.1371/journal.pone.0177680.s001

(DOCX)

Acknowledgments

This work was conducted with support from the Harvard NeuroDiscovery Center, Muscular Dystrophy Association Clinical Research Training Grant, Research fellowship from the American Academy of Neurology, and the Anne B. Young neuroscience translational medicine fellowship. We like to thank Drs. Dinesh Deelchand and Gulin Oz from the Center for Magnetic Resonance Research at the University of Minnesota for providing us with the 7T-LCModel basis-set. In addition, we like to thank Dr. Merit Cudkowicz for insightful discussion on the study design. Lastly, we like to thank all participating patients and their families.

Author Contributions

  1. Conceptualization: NA BR EMR.
  2. Data curation: NA MX EMR.
  3. Formal analysis: MX CJL AM EMR.
  4. Funding acquisition: NA EMR.
  5. Investigation: NA MX CT BK RL PC SP AM NS RS EMR.
  6. Methodology: NA MX CT BK CJL SP NS RS BR EMR.
  7. Project administration: NA EMR.
  8. Resources: NA SP.
  9. Software: CT BK CJL NS RS EMR.
  10. Supervision: NA BR EMR.
  11. Validation: MX EMR.
  12. Visualization: MX EMR.
  13. Writing – original draft: NA ER EMR.
  14. Writing – review & editing: NA MX CT BK ER CJL SP NS RS BR EMR.

References

  1. 1. Turner MR, Hardiman O, Benatar M, Brooks BR, Chio A, de Carvalho M, et al. Controversies and priorities in amyotrophic lateral sclerosis. Lancet Neurol. 2013;12(3):310–22. pmid:23415570
  2. 2. Turner MR, Bowser R, Bruijn L, Dupuis L, Ludolph A, McGrath M, et al. Mechanisms, models and biomarkers in amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener. 2013;14 Suppl 1:19–32.
  3. 3. Rothstein JD, Kuncl RW. Neuroprotective strategies in a model of chronic glutamate-mediated motor neuron toxicity. J Neurochem. 1995;65(2):643–51. pmid:7616220
  4. 4. Evans MC, Couch Y, Sibson N, Turner MR. Inflammation and neurovascular changes in amyotrophic lateral sclerosis. Mol Cell Neurosci. 2013;53:34–41. pmid:23110760
  5. 5. Foerster BR, Welsh RC, Feldman EL. 25 years of neuroimaging in amyotrophic lateral sclerosis. Nat Rev Neurol. 2013;9(9):513–24. pmid:23917850
  6. 6. Turner MR, Modo M. Advances in the application of MRI to amyotrophic lateral sclerosis. Expert Opin Med Diagn. 2010;4(6):483–96. pmid:21516259
  7. 7. Rocha AJ, Maia AC Junior. Is magnetic resonance imaging a plausible biomarker for upper motor neuron degeneration in amyotrophic lateral sclerosis/primary lateral sclerosis or merely a useful paraclinical tool to exclude mimic syndromes? A critical review of imaging applicability in clinical routine. Arq Neuropsiquiatr. 2012;70(7):532–9. pmid:22836461
  8. 8. Wang S, Melhem ER, Poptani H, Woo JH. Neuroimaging in amyotrophic lateral sclerosis. Neurotherapeutics. 2011;8(1):63–71. pmid:21274686
  9. 9. Kaufmann P, Mitsumoto H. Amyotrophic lateral sclerosis: objective upper motor neuron markers. Curr Neurol Neurosci Rep. 2002;2(1):55–60. pmid:11898584
  10. 10. Bradley WG, Bowen BC, Pattany PM, Rotta F. 1H-magnetic resonance spectroscopy in amyotrophic lateral sclerosis. J Neurol Sci. 1999;169(1–2):84–6. pmid:10540013
  11. 11. Pohl C, Block W, Karitzky J, Traber F, Schmidt S, Grothe C, et al. Proton magnetic resonance spectroscopy of the motor cortex in 70 patients with amyotrophic lateral sclerosis. Arch Neurol. 2001;58(5):729–35. pmid:11346367
  12. 12. Mitsumoto H, Ulug AM, Pullman SL, Gooch CL, Chan S, Tang MX, et al. Quantitative objective markers for upper and lower motor neuron dysfunction in ALS. Neurology. 2007;68(17):1402–10. pmid:17452585
  13. 13. Pioro EP. MR spectroscopy in amyotrophic lateral sclerosis/motor neuron disease. J Neurol Sci. 1997;152 Suppl 1:S49–53.
  14. 14. Pioro EP, Majors AW, Mitsumoto H, Nelson DR, Ng TC. 1H-MRS evidence of neurodegeneration and excess glutamate + glutamine in ALS medulla. Neurology. 1999;53(1):71–9. pmid:10408539
  15. 15. Bowen BC, Pattany PM, Bradley WG, Murdoch JB, Rotta F, Younis AA, et al. MR imaging and localized proton spectroscopy of the precentral gyrus in amyotrophic lateral sclerosis. AJNR Am J Neuroradiol. 2000;21(4):647–58. pmid:10782773
  16. 16. Schuff N, Rooney WD, Miller R, Gelinas DF, Amend DL, Maudsley AA, et al. Reanalysis of multislice (1)H MRSI in amyotrophic lateral sclerosis. Magn Reson Med. 2001;45(3):513–6. pmid:11241711
  17. 17. Wang S, Poptani H, Woo JH, Desiderio LM, Elman LB, McCluskey LF, et al. Amyotrophic lateral sclerosis: diffusion-tensor and chemical shift MR imaging at 3.0 T. Radiology. 2006;239(3):831–8. pmid:16641339
  18. 18. Verma G, Woo JH, Chawla S, Wang S, Sheriff S, Elman LB, et al. Whole-brain analysis of amyotrophic lateral sclerosis by using echo-planar spectroscopic imaging. Radiology. 2013;267(3):851–7. pmid:23360740
  19. 19. Sivak S, Bittsansky M, Kurca E, Turcanova-Koprusakova M, Grofik M, Nosal V, et al. Proton magnetic resonance spectroscopy in patients with early stages of amyotrophic lateral sclerosis. Neuroradiology. 2010;52(12):1079–85. pmid:20369234
  20. 20. Unrath A, Ludolph AC, Kassubek J. Brain metabolites in definite amyotrophic lateral sclerosis. A longitudinal proton magnetic resonance spectroscopy study. J Neurol. 2007;254(8):1099–106. pmid:17431700
  21. 21. Rule RR, Suhy J, Schuff N, Gelinas DF, Miller RG, Weiner MW. Reduced NAA in motor and non-motor brain regions in amyotrophic lateral sclerosis: a cross-sectional and longitudinal study. Amyotroph Lateral Scler Other Motor Neuron Disord. 2004;5(3):141–9. pmid:15512902
  22. 22. Lombardo F, Frijia F, Bongioanni P, Canapicchi R, Minichilli F, Bianchi F, et al. Diffusion tensor MRI and MR spectroscopy in long lasting upper motor neuron involvement in amyotrophic lateral sclerosis. Arch Ital Biol. 2009;147(3):69–82. pmid:20014653
  23. 23. Govind V, Sharma KR, Maudsley AA, Arheart KL, Saigal G, Sheriff S. Comprehensive evaluation of corticospinal tract metabolites in amyotrophic lateral sclerosis using whole-brain 1H MR spectroscopy. PLoS One. 2012;7(4):e35607. pmid:22539984
  24. 24. Kalra S, Vitale A, Cashman NR, Genge A, Arnold DL. Cerebral degeneration predicts survival in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2006;77(11):1253–5. pmid:16835288
  25. 25. Abe K, Takanashi M, Watanabe Y, Tanaka H, Fujita N, Hirabuki N, et al. Decrease in N-acetylaspartate/creatine ratio in the motor area and the frontal lobe in amyotrophic lateral sclerosis. Neuroradiology. 2001;43(7):537–41. pmid:11512581
  26. 26. Liu C, Jiang R, Yi X, Zhu W, Bu B. Role of diffusion tensor imaging or magnetic resonance spectroscopy in the diagnosis and disability assessment of amyotrophic lateral sclerosis. J Neurol Sci. 2015;348(1–2):206–10. pmid:25524526
  27. 27. Kalra S, Hanstock CC, Martin WR, Allen PS, Johnston WS. Detection of cerebral degeneration in amyotrophic lateral sclerosis using high-field magnetic resonance spectroscopy. Arch Neurol. 2006;63(8):1144–8. pmid:16908742
  28. 28. Block W, Karitzky J, Traber F, Pohl C, Keller E, Mundegar RR, et al. Proton magnetic resonance spectroscopy of the primary motor cortex in patients with motor neuron disease: subgroup analysis and follow-up measurements. Arch Neurol. 1998;55(7):931–6. pmid:9678310
  29. 29. van der Graaff MM, Lavini C, Akkerman EM, Majoie Ch B, Nederveen AJ, Zwinderman AH, et al. MR spectroscopy findings in early stages of motor neuron disease. AJNR Am J Neuroradiol. 2010;31(10):1799–806. pmid:20801763
  30. 30. Han J, Ma L. Study of the features of proton MR spectroscopy ((1)H-MRS) on amyotrophic lateral sclerosis. J Magn Reson Imaging. 2010;31(2):305–8. pmid:20099342
  31. 31. Cedarbaum JM, Stambler N, Malta E, Fuller C, Hilt D, Thurmond B, et al. The ALSFRS-R: a revised ALS functional rating scale that incorporates assessments of respiratory function. BDNF ALS Study Group (Phase III). J Neurol Sci. 1999;169(1–2):13–21. pmid:10540002
  32. 32. Kimura F, Fujimura C, Ishida S, Nakajima H, Furutama D, Uehara H, et al. Progression rate of ALSFRS-R at time of diagnosis predicts survival time in ALS. Neurology. 2006;66(2):265–7. pmid:16434671
  33. 33. Tkac I, Starcuk Z, Choi IY, Gruetter R. In vivo 1H NMR spectroscopy of rat brain at 1 ms echo time. Magn Reson Med. 1999;41(4):649–56. pmid:10332839
  34. 34. Tkac I, Andersen P, Adriany G, Merkle H, Ugurbil K, Gruetter R. In vivo 1H NMR spectroscopy of the human brain at 7 T. Magn Reson Med. 2001;46(3):451–6. pmid:11550235
  35. 35. Govindaraju V, Young K, Maudsley AA. Proton NMR chemical shifts and coupling constants for brain metabolites. NMR Biomed. 2000;13(3):129–53. pmid:10861994
  36. 36. Tkac I, Oz G, Adriany G, Ugurbil K, Gruetter R. In vivo 1H NMR spectroscopy of the human brain at high magnetic fields: metabolite quantification at 4T vs. 7T. Magn Reson Med. 2009;62(4):868–79. pmid:19591201
  37. 37. Deelchand DK, Henry PG, Ugurbil K, Marjanska M. Measurement of transverse relaxation times of J-coupled metabolites in the human visual cortex at 4 T. Magn Reson Med. 2012;67(4):891–7. pmid:21748799
  38. 38. Foerster BR, Pomper MG, Callaghan BC, Petrou M, Edden RA, Mohamed MA, et al. An imbalance between excitatory and inhibitory neurotransmitters in amyotrophic lateral sclerosis revealed by use of 3-T proton magnetic resonance spectroscopy. JAMA Neurol. 2013;70(8):1009–16. pmid:23797905
  39. 39. Urenjak J, Williams SR, Gadian DG, Noble M. Proton nuclear magnetic resonance spectroscopy unambiguously identifies different neural cell types. J Neurosci. 1993;13(3):981–9. pmid:8441018
  40. 40. Weiduschat N, Mao X, Hupf J, Armstrong N, Kang G, Lange DJ, et al. Motor cortex glutathione deficit in ALS measured in vivo with the J-editing technique. Neurosci Lett. 2014;570:102–7. pmid:24769125
  41. 41. Kaufmann P, Pullman SL, Shungu DC, Chan S, Hays AP, Del Bene ML, et al. Objective tests for upper motor neuron involvement in amyotrophic lateral sclerosis (ALS). Neurology. 2004;62(10):1753–7. pmid:15159473
  42. 42. Sudharshan N, Hanstock C, Hui B, Pyra T, Johnston W, Kalra S. Degeneration of the mid-cingulate cortex in amyotrophic lateral sclerosis detected in vivo with MR spectroscopy. AJNR Am J Neuroradiol. 2011;32(2):403–7. pmid:21087934
  43. 43. Choi DW. Glutamate neurotoxicity and diseases of the nervous system. Neuron. 1988;1(8):623–34. pmid:2908446
  44. 44. Cleveland DW, Rothstein JD. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat Rev Neurosci. 2001;2(11):806–19. pmid:11715057
  45. 45. Brown AM, Ransom BR. Astrocyte glycogen and brain energy metabolism. Glia. 2007;55(12):1263–71. pmid:17659525
  46. 46. Rothstein JD, Tsai G, Kuncl RW, Clawson L, Cornblath DR, Drachman DB, et al. Abnormal excitatory amino acid metabolism in amyotrophic lateral sclerosis. Ann Neurol. 1990;28(1):18–25. pmid:2375630
  47. 47. Danbolt NC. Glutamate uptake. Prog Neurobiol. 2001;65(1):1–105. pmid:11369436
  48. 48. Daikhin Y, Yudkoff M. Compartmentation of brain glutamate metabolism in neurons and glia. J Nutr. 2000;130(4S Suppl):1026S–31S. pmid:10736375
  49. 49. Duarte JM, Lei H, Mlynarik V, Gruetter R. The neurochemical profile quantified by in vivo 1H NMR spectroscopy. Neuroimage. 2012;61(2):342–62. pmid:22227137
  50. 50. Tiret B, Shestov AA, Valette J, Henry PG. Metabolic Modeling of Dynamic (13)C NMR Isotopomer Data in the Brain In Vivo: Fast Screening of Metabolic Models Using Automated Generation of Differential Equations. Neurochem Res. 2015;40(12):2482–92. pmid:26553273
  51. 51. Griffith HR, Okonkwo OC, O'Brien T, Hollander JA. Reduced brain glutamate in patients with Parkinson's disease. NMR Biomed. 2008;21(4):381–7. pmid:17907264
  52. 52. Shiino A, Watanabe T, Shirakashi Y, Kotani E, Yoshimura M, Morikawa S, et al. The profile of hippocampal metabolites differs between Alzheimer's disease and subcortical ischemic vascular dementia, as measured by proton magnetic resonance spectroscopy. J Cereb Blood Flow Metab. 2012;32(5):805–15. pmid:22314267
  53. 53. Sargsyan SA, Monk PN, Shaw PJ. Microglia as potential contributors to motor neuron injury in amyotrophic lateral sclerosis. Glia. 2005;51(4):241–53. pmid:15846792
  54. 54. Kawamata T, Akiyama H, Yamada T, McGeer PL. Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am J Pathol. 1992;140(3):691–707. pmid:1347673
  55. 55. Henkel JS, Engelhardt JI, Siklos L, Simpson EP, Kim SH, Pan T, et al. Presence of dendritic cells, MCP-1, and activated microglia/macrophages in amyotrophic lateral sclerosis spinal cord tissue. Ann Neurol. 2004;55(2):221–35. pmid:14755726
  56. 56. Brettschneider J, Toledo JB, Van Deerlin VM, Elman L, McCluskey L, Lee VM, et al. Microglial activation correlates with disease progression and upper motor neuron clinical symptoms in amyotrophic lateral sclerosis. PLoS One. 2012;7(6):e39216. pmid:22720079
  57. 57. Dangond F, Hwang D, Camelo S, Pasinelli P, Frosch MP, Stephanopoulos G, et al. Molecular signature of late-stage human ALS revealed by expression profiling of postmortem spinal cord gray matter. Physiol Genomics. 2004;16(2):229–39. pmid:14645737
  58. 58. Zurcher NR, Loggia ML, Lawson R, Chonde DB, Izquierdo-Garcia D, Yasek JE, et al. Increased in vivo glial activation in patients with amyotrophic lateral sclerosis: assessed with [(11)C]-PBR28. Neuroimage Clin. 2015;7:409–14. pmid:25685708