Dikoma C. Shungu1, Kristin Engelstadt2, Xiangling Mao1, Guoxin Kang1, Aya Goji1, Robert H. Fryer2, Savalatore DiMauro2, and Darryl C. De Vivo2
1Radiology, Weill Cornell Medical College, New York, NY, United States, 2Neurology, College of Physicians and Surgeons of Columbia University, New York, NY, United States
Synopsis
Although mitochondrial dysfunction has been associated with redox dysregulation, in vivo human brain evidence of such an association is currently lacking. This study aimed to use 1H MRS to measure brain levels of the primary tissue antioxidant glutathione (GSH) in patients with MELAS – a primary mitochondrial disorder – as an objective marker of CNS oxidative stress in such disorders. Compared to healthy control subjects, patients with MELAS showed a 31% lower cortical GSH levels, thereby directly implicating CNS oxidative stress as a player in the disorder and pointing to potential therapeutic interventions based on elevating the levels of cerebral antioxidants.
The syndrome of mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes (MELAS) is a devastating and invariably fatal neurological disorder due to an A to G transition at nucleotide 3243 of the mitochondria DNA (mtDNA) [1]. While impairments of mitochondrial function and energy production that are associated with the mtDNA mutation in MELAS have been extensively investigated, there have been comparatively fewer studies of one of the most profound consequences of mitochondrial dysfunction, which is redox imbalance or oxidative stress due to increased production and accumulation of free radicals. In the present study, we aimed to begin filling this knowledge gap by using 1H MRS to measure levels of glutathione (GSH) – the most abundant and important antioxidant in living tissue – in vivo in the brain of MELAS patients as an objective marker of redox imbalance and oxidative stress in the disorder.
Subjects: Participants in this study consisted of 6 patients with MELAS and 10 medically healthy subjects, who served as the normal comparison group. The diagnosis of MELAS was based on documentation of the A3243G mtDNA mutation, as well as on the clinical definition and manifestations of the disorder, which include encephalopathy, with seizures; stroke-like episodes; and evidence for mitochondrial defects, such as lactic acidosis or ragged-red fibers in the muscle biopsy.
Brain GSH Measurements In Vivo by J-edited 1H MRS: In vivo cortical GSH spectra were recorded on a 3.0 T GE MR system from a 2.5x2.5x4.0-cm3 voxel prescribed in the occipital cortex. As MELAS diffusely affected the brain, our primary considerations in selecting the occipital cortex and a relatively large voxel of interest were to maximize the quality of the GSH spectra through achievement of good static field homogeneity and high signal-to-noise ratios. Each spectrum was acquired in 15 min using the standard J-edited spin echo difference technique (Figure 1) and an 8-channel phased-array head coil, with TE/TR 68/1500 ms and 290 interleaved excitations (580 total). Levels of GSH in the voxel of interest were derived by frequency-domain fitting of the GSH peak in the difference edited spectra (Figure 1), and then expressing the resulting peak areas as ratios relative to the area of the unsuppressed tissue water (W) signal in the same voxel. General linear model analysis was conducted to compare GSH/W values in the MELAS patients with those in the healthy control subjects.
As shown in Figure 2, mean occipital GSH/W in the MELAS patients ([1.5±0.3]x10-3) was significantly lower (p < 0.002) than in the healthy control subjects (2.2±0.4]x10-3) – a 31% lower mean GSH/W in MELAS with a robust effect size (Cohen’s d) of 1.9 that translates into an overlap between the two distributions of just 20.6%, despite the relatively small sample size.
DISCUSSION AND CONCLUSION
To our knowledge, this preliminary study is the first to report levels of GSH in the brain of living MELAS patients, finding them to be robustly lower than in healthy controls, and thereby directly implicating CNS oxidative stress as a player in the disorder. Our search of the literature found only one prior study, in which GSH levels were measured in whole blood from patients with various mitochondrial disorders, including 8 patients with MELAS [2]. While that study reported GSH to be lower in patients with mitochondrial disorders as a group, it failed to find a GSH deficit specifically in whole blood of MELAS patients. The disagreement between this single in vitro measure of GSH with our in vivo measure reported here for MELAS patients suggests the importance of performing such measurements directly on living brain, since (a) the antioxidant does not readily cross the blood-brain barrier (BBB) [3] and (b) we have found the mean in vivo brain levels to be lower in MELAS than in controls with a large effect size for a comparable number of subjects. Lastly, the present finding of a significant GSH deficit in MELAS patients in vivo, coupled with its poor BBB permeability, has therapeutic implications. It provides a compelling rationale to investigate synthetic GSH precursors (e.g., N-acetylcysteine or NAC) that can cross the BBB and can be administered to spur in situ synthesis and elevations of cortical GSH levels thereby promoting neuroprotection against oxidative stress [4].
Acknowledgements
This study was supported by NIH grant P01HD080642.References
[1] Hirano, M., Pavlakis, S.G., 1994. J. Child Neurol. 9, 4–13;
[2] Enns, Gregory M., et al. PloS one 9.6 (2014): e100001.
[3] Martin, Heather L., and Peter Teismann. The FASEB Journal 23.10 (2009): 3263-3272.
[4] Atkuri, Kondala R., et al. Current opinion in pharmacology 7.4 (2007): 355-359.