Free Radical Biology and Medicine 93 (2016) 110–117

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

Do glutathione levels decline in aging human brain? Junchao Tong a,b,n, Paul S. Fitzmaurice c, Anna Moszczynska d, Katie Mattina a,b, Lee-Cyn Ang e, Isabelle Boileau b, Yoshiaki Furukawa f, Napapon Sailasuta g, Stephen J. Kish a a

Human Brain Laboratory, Research Imaging Centre, Centre for Addiction and Mental Health, 250 College Street, Toronto, Ontario, Canada M5T 1R8 Addiction Imaging Research Group, Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, Canada c The Drug Detection Agency, Auckland, New Zealand d Department of Pharmaceutical Sciences, Eugene Applebaum College of Pharmacy and Health Sciences, Wayne State University, Detroit, MI, USA e Division of Neuropathology, London Health Sciences Centre, University of Western Ontario, London, Ontario, Canada f Department of Neurology, Juntendo Tokyo Koto Geriatric Medical Center, and Faculty of Medicine, University and Post Graduate University of Juntendo, Tokyo, Japan g Research Imaging Centre, Centre for Addiction and Mental Health, Toronto, Ontario, Canada b

art ic l e i nf o

a b s t r a c t

Article history: Received 15 December 2015 Received in revised form 28 January 2016 Accepted 31 January 2016 Available online 1 February 2016

For the past 60 years a major theory of “aging” is that age-related damage is largely caused by excessive uncompensated oxidative stress. The ubiquitous tripeptide glutathione is a major antioxidant defense mechanism against reactive free radicals and has also served as a marker of changes in oxidative stress. Some (albeit conflicting) animal data suggest a loss of glutathione in brain senescence, which might compromise the ability of the aging brain to meet the demands of oxidative stress. Our objective was to establish whether advancing age is associated with glutathione deficiency in human brain. We measured reduced glutathione (GSH) levels in multiple regions of autopsied brain of normal subjects (n ¼74) aged one day to 99 years. Brain GSH levels during the infancy/teenage years were generally similar to those in the oldest examined adult group (76–99 years). During adulthood (23–99 years) GSH levels remained either stable (occipital cortex) or increased (caudate nucleus, frontal and cerebellar cortices). To the extent that GSH levels represent glutathione antioxidant capacity, our postmortem data suggest that human brain aging is not associated with declining glutathione status. We suggest that aged healthy human brains can maintain antioxidant capacity related to glutathione and that an age-related increase in GSH levels in some brain regions might possibly be a compensatory response to increased oxidative stress. Since our findings, although suggestive, suffer from the generic limitations of all postmortem brain studies, we also suggest the need for “replication” investigations employing the new 1H MRS imaging procedures in living human brain. & 2016 Elsevier Inc. All rights reserved.

Keywords: Glutathione Oxidative stress Aging Human brain Postmortem

1. Introduction Oxidative stress is considered a hallmark of human aging [1–4]; accordingly a variety of antioxidants have been avidly sought as possible dietary supplements or therapeutic reagents for the purpose of anti-aging and protection from degenerative aging changes [5,6]. The antioxidant glutathione (γ-L-glutamyl-L-cysteinylglycine, GSH, the reduced form), which is synthesized de novo by γ-glutamyl cysteine ligase and glutathione synthetase, plays a Abbreviations: GSH, Reduced glutathione; GSSG, Oxidized glutathione; MRS, Magnetic resonance spectroscopy; NSE, Neuron specific enolase; PMI, Postmortem interval n Corresponding author at: Human Brain Lab, Research Imaging Centre, Centre for Addiction and Mental Health, 250 College Street, Toronto, Ontario, Canada M5T 1R8. E-mail address: [email protected] (J. Tong). http://dx.doi.org/10.1016/j.freeradbiomed.2016.01.029 0891-5849/& 2016 Elsevier Inc. All rights reserved.

pivotal role in supporting oxidative defense and maintaining redox equilibrium of the brain [7]. Glutathione participates in elimination of xenobiotics and hydrogen/organic peroxides through the actions of glutathione S-transferase and glutathione peroxidase, respectively, and can be recycled from the oxidized form (GSSG) to the reduced form (GSH) by glutathione reductase. Below normal levels of GSH have been reported in autopsied brain of patients with some neurodegenerative disorders [8–12] and it has been suggested that an age-related decline in glutathione in the normal human might “…contribute to many of the age-related declines in cellular function as well as the increased susceptibility to various insults” (see [13], p. 309). This glutathione-aging hypothesis has raised the interesting possibility that supplementation of GSH in the forms of its precursors, e.g., N-acetylcysteine, might be helpful in slowing down the progression of aging-related human disorders [14–16]. The cornerstone of the “GSH deficiency” hypothesis of brain

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

aging is the preclinical observations of decreased levels of GSH in peripheral organs [17] and brain [18–20] of senescent as compared to mature animals. However, a review of the preclinical literature emphasized the lack of consensus in reports on aging- and development-related changes in brain levels of GSH, sometimes even in the same strain, sex, and the brain regions examined (see Section 4.2). There is also little information regarding changes in levels of GSH in human brain aging with the exception of a preliminary proton magnetic resonance spectroscopy (1H MRS) report of decreased GSH levels in the occipital lobe of elderly versus young subjects [21] and a postmortem observation of decreased hippocampal GSH levels with aging in brain of a cohort of subjects suffering from head trauma ([22,23]; see Section 4.2). Given the uncertain state of the literature on glutathione and brain aging in humans and the potential clinical therapeutic relevance of this question, the aim of our study was to establish whether levels of reduced and oxidized GSH change in developing and aging postmortem human brain. For this purpose, a representative number (n ¼74) of autopsied human brain from infancy to senescence was employed. Alterations in GSH levels across brain areas were explored for region-specific changes and we hypothesized, based on some, albeit conflicting, animal literature, that levels of GSH might decline with advancing age.

111

2. Materials and methods This study was approved by the Centre for Addiction and Mental Health Research Ethics Board. Brain tissues were obtained at necropsy from a total of 74 subjects (male, n¼ 44; female, n¼30) who died without evidence of neurological or psychiatric disease (see Table 1 for subject information and known or suspected cause of death) or brain pathology when the fixed half brain was used for neuropathological examination [postmortem interval (PMI, from 3 to 27 h), 12.970.8 h, mean7SEM]. The agonal status of the subjects (Table 1) was classified according to their mode of death following the criteria of Hardy [24]: 1, violent fast death; 2, fast death of natural causes; 3, intermediate unexpected death; and 4, slow death with prolonged terminal phase. At necropsy, each brain was removed and divided midsagitally. One half of the brain (left, n¼43; right, n¼31) was frozen at  80 °C until dissection for biochemical analysis. The brain regions were dissected as previously described [25] and using the Atlas of Riley [26] for the caudate and Brodmann classification for cerebral cortical brain areas (frontal cortex, BA9; occipital cortex, BA17). Levels of GSH and GSSG were measured by a coulometric method using high performance liquid chromatography (HPLC; Spectrophysics SP8800 HPLC pump, ThermoFinnigan) and electrochemical detection (ESA Coulochem 2; ESA, Chelmsford, MA) with coulometric cells (ESA guard cell 5020 and analytical cell 5011) as previously described [12]. Briefly,

Table 1 Subject informationa. Case#/sex/ side

Ageb

PMI (h)

Agonalc

Probable cause of death

Case#/sex/ side

Ageb

PMI (h)

Agonalc

Probable cause of death

1/F/L 2/F/L 3/F/R 4/M/L 5/M/R 6/F/L 7/M/L 8/F/L 9/M/L 10/M/R 11/M 12/F/L 13/F/R 14/M/L 15/M/R 16/M/L 17/M/R 18/M/L 19/F/L 20/M/L 21/F/L 22/M/L 23/F/L 24/M/L 25/F/L 26/M/L 27/M/L 28/M/R 29/M/L 30/M/L 31/F/L 32/M/R 33/F/L 34/M/R 35/M/L 36/M/L # 37/M/L

21 h 8d 9d 9d 21 d 24 d 26 d 28 d 35 d 2m 2m 3.5 m 4m 4.5 m 6m 6.5 m 7m 7m 8m 8m 10 m 11 m 1.5 1.7 2 2 2.7 3.5 4 7 11 13 14 17 18 23 24

20 9 8 10 3 23 5 14 3.5 18 10 9 12 10 6 16 15 26 18 15 21 12 7 16 6 10 7 11 9 22 17 27 12 6.5 16.5 25 5.25

3 2 2 2 – 4 3 3 2 2 2 2 3 2 2 3 2 – 3 1 2 2 3 2 1 2 3 1 3 3 2 1 3 1 1 4 1

Birth asphyxia Hypoplastic left heart Hypoplastic left heart Pulmonary atresia Unknown Asphyxia TAPVD Gastroenteritis Nesidioblastosis CHF Cardiomyopathy Myocarditis Sepsis Primary hypertension Coarctation Pneumonia Myocarditis Unknown Gastroenteritis Acute trauma abdomen Drowning Drowning Pulmonary atresia Post repair for TOF Abdominal trauma Unexpected death Small bowel obstruction Chest trauma Chronic heart disease Acute adrenal insufficiency Congenital heart disease Gunshot to heart Renal failure Abdominal trauma Multiple trauma accident Morbid obesity Multiple trauma accident

38/M/R 39/F/L *40/M/R 41/M/R 42/M/R 43/M/L 44/M/R 45/M/R 46/M/R # 47/M/L # 48/F/R 49/M/L 50/F/R 51/F/R 52/F/L 53/F/L 54/F/L 55/M/R 56/F/R 57/M/R 58/M/R 59/M/R 60/F/R 61/M/L 62/F/R 63/M/R 64/F/L 65/F/R 66/F/R 67/M/R 68/F/L 69/M/L 70/M/L 71/M/L 72/F/L 73/F/R 74/F/L

28 28 31 36 37 38 41 44 47 48 48 50 55 58 59 60 66 69 70 71 71 72 72 73 73 76 77 80 80 80 83 83 86 87 89 92 99

17 7 13 20.5 10.3 8.5 5 13 23 5.25 22.3 10.3 20.5 13 21.5 18.5 3 12 9.5 15 17.5 8.5 24.8 16.3 7.5 4 11 10 10.5 15 3 22 9.5 12 8 3.5 24

2 1 2 1 2 2 3 2 2 2 2 2 4 4 2 4 2 4 2 2 2 2 2 2 4 3 2 3 2 2 2 2 3 4 2 – –

Pulmonary embolism Homicide hemorrhage Cardiomyopathy Multiple trauma accident Pulmonary embolism ASCVD ASCVD ASCVD ASCVD Cardiomyopathy ASCVD HASCVD Bronchopneumonia Pulmonary edema HASCVD Bronchial hemorrhage Pulmonary embolism Renal failure/pneumonia MI, ASCAD MI, CHF Natural death MI Cardiomyopathy MI Pulmonary consolidation, breast cancer Exsanguination MI Atheroembolization MI HASCVD MI CHF Hypertension, coronary disease, arrhythmia Diffuse interstial pulmonary disease CHF Unknown Unknown

a

Subjects are Caucasians unless otherwise indicated (*Asian; #African American). Ages are in years unless otherwise indicated (h, hours; d, days; m, months). c Agonal status of subjects was classified according to the criteria of Hardy et al. (1985) [24] [1, violent fast death; 2, natural fast death; 3, intermediate unexpected death; 4, slow death]. M, male; F, female; L, left side; R, right side; PMI, postmortem interval; ASCAD, atherosclerotic coronary artery disease; ASCVD, arteriosclerotic cardiovascular disease; CHF, congestive heart failure; HASCVD, hypertensive ASCVD; MI, myocardial infarction; TAPVD, total anomalous pulmonary venous drainage; TOF, Tetralogy of Fallot. b

112

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

chromatography was performed on a reverse-phase ODS spherosorb column (250  4.7 mm2, 5 mm; Hichrom, Berkshire, UK). The eluent consisted of NaH2PO4 (10 mM, adjusted to pH 2.7 using phosphoric acid), 0.05 mM octane sulfonate, and 6.9% methanol at a flow rate of 1 mL/min. Brain tissue (30–40 mg) was homogenized on ice in 10 volume NaH2PO4 (10 mM, pH2.7) containing 0.05 mM octane sulfonate. An equal volume of ice-cold methanol was added to precipitate the proteins. After centrifugation at 17,000g for 15 min, the supernatant was diluted with five volumes of homogenization solution and filtered (0.25 mm, syringe filter) before injection. All procedures were

performed on ice or at 0–4 °C to minimize GSH auto-oxidation and degradation. Protein concentration was determined using the Bio-Rad Protein Assay Kit (Bio-Rad, Hercules, CA, USA) with bovine plasma albumin as the standard. Levels of the housekeeping control protein neuron specific enolase (NSE) in the frontal cortex of the same subjects were previously published [27]. Statistical analysis of the data (StatSoft STATISTICA 7.1, Tulsa, Oklahoma, USA) was conducted using two different approaches. First, the 74 subjects were subdivided into six age groups [age, mean 7SEM (range, n); see Table 2 and Table 3]: A1, 26 76 days

Table 2 Levels of reduced glutathione (GSH) in human brain development and aging. Age group Developing brain A1: 21 h–2 months A2: 3 months–1 year A3: 1 year–18 years F (A1–A3) r (21 h–18 years) r (21 h–1 year) r (1 year–18 years) Adult brain A4: 23 years–50 years A5: 55 years–75 years A6: 76 yrs–99 years F (A4–A6) r (23 years–99 years) r (23 year–50 years)

Age mean

Caudate

Frontal cortex

Occipital cortex

Cerebellar cortex

26 days 7 months 8 years

8.87 7 0.64 6.95 7 0.80 5.66 70.58a,c 5.86*  0.35*  0.47*  0.11

7.65 7 0.80 5.30 7 0.79 3.46 7 0.46a,c,e 9.75*  0.50*  0.52*  0.50

7.147 0.67 5.78 7 0.56 5.7070.77 1.39 0.01  0.37 0.28

4.487 0.27 4.99 7 0.69 4.86 7 0.44 0.30 0.08  0.06 0.13

37 years 67 years 84 years

5.36 7 0.70c 6.58 7 0.54 7.75 7 0.80 2.93 0.34* 0.10

3.187 0.49c,e 7.45 7 0.94b 6.89 7 1.18b 7.64* 0.45* 0.21

5.50 7 0.58 5.78 7 0.40 6.22 7 0.50 0.50 0.14 0.27

2.80 7 0.27d 3.577 0.30 3.6770.50 1.94 0.47* 0.16

3.68*  0.02

6.63* 0.17

1.00  0.05

4.42*  0.30*

Overall F (A1–A6) r (21 h–99 years)

Data (mean7 SEM) in μg/mg protein of n¼ 9–14 for each age group. *p o0.05 for one-way ANOVA (F) or Pearson product-moment correlation (r). Post hoc analyzes with Bonferroni adjustment: a

p o 0.05, A3 vs. A1 among the developing brain groups (A1–A3); p o0.05, A5 or A6 vs. A4 among the adult brain groups (A4–A6); po 0.05, A3 or A4 vs. A1 among all groups (A1–A6); d po 0.05, A4 vs. A2 or A3 among all groups (A1–A6); e p o 0.05, A3 or A4 vs. A5 or A6 among all groups (A1–A6). b c

Table 3 Levels of oxidized glutathione (GSSG) in human brain development and aging. Age group Developing brain A1: 21 h–2 months A2: 3 months–1 year A3: 1 year–18 years F (A1–A3) r (21 h–18 years) r (21 h–1 year) r (1 year–18 years) Adult brain A4: 23 years–50 years A5: 55 years–75 years A6: 76 years–99 years F (A4–A6) r (23 years–99 years) r (23 year–50 years) Overall F (A1–A6) r (21 h–99 years)

Age mean

Caudate

Frontal cortex

Occipital cortex

Cerebellar cortex

26 days 7 months 8 years

0.497 0.06 0.43 7 0.07 0.477 0.08 0.14  0.11  0.35  0.21

0.93 7 0.16 0.54 7 0.16 0.54 7 0.14 2.09  0.32  0.51*  0.44

0.36 7 0.03 0.197 0.04a 0.25 7 0.06 3.65* 0.07  0.75* 0.33

0.137 0.06 0.157 0.05 0.08 7 0.03 0.72  0.15 0.06  0.07

37 years 67 years 84 years

0.26 7 0.06 0.317 0.05 0.52 7 0.08b 4.36* 0.44* 0.13

0.22 7 0.05c 0.31 70.08c 0.39 7 0.11c 1.05 0.17 0.19

0.167 0.04c 0.22 7 0.06 0.187 0.05 0.46 0.15 0.28

0.167 0.05 0.107 0.03 0.217 0.09 0.94 0.12 0.33

2.23  0.05

4.45*  0.35*

2.53*  0.18

0.80 0.12

Data (mean7 SEM) in μg/mg protein of n¼ 9–14 for each age group. *p o0.05 for one-way ANOVA (F) or Pearson product-moment correlation (r). Post hoc analyzes with Bonferroni adjustment: a b c

p o 0.05, A2 vs. A1 among the developing brain groups (A1–A3); p o0.05, A6 vs. A4 among the adult brain groups (A4–A6); po 0.05, A3, A4, A5 or A6 vs. A1 among all groups (A1–A6).

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

113

(21 h to 61 days, n ¼11); A2, 6.9 70.7 months (3.5–11 months, n ¼11); A3, 7.5 71.7 years (1.5–18 years, n ¼13); A4, 37.572.5 years (23–50 years, n ¼14); A5, 66.8 71.8 years (55–75 years, n ¼13); and A6, 84.3 71.9 years (76–99 years, n ¼12). One-way ANOVA for the overall six groups and separate one-way ANOVAs for the three youngest and the three oldest age groups, followed by post hoc Bonferroni adjustment, were performed. Secondly, Pearson product-moment correlation was used to evaluate the effect of age on levels of GSH and GSSG in the full age range (21 h to 99 years) and in separate age groups (see Tables 2 and 3). An alpha of p o0.05 was used as the criterion for statistical significance.

3. Results 3.1. Overall changes of GSH from birth to senescence Levels of GSH were not significantly correlated with age from infancy (21 h) to 99 years of age in brain regions examined with the exception of a slight negative correlation in the cerebellar cortex (r ¼  0.30, n ¼67, p ¼0.013; see Table 2). One way ANOVA disclosed differences amongst the six subdivided age groups in caudate (F5,64 ¼ 3.68, p ¼0.005), frontal cortex (F5,64 ¼6.63, p o0.0001), and cerebellar cortex (F5,61 ¼4.42, p ¼0.002) but not in occipital cortex (F5,63 ¼1.00, p ¼0.43), suggesting differential changes during development and aging in these brain regions. However, no significant difference in GSH levels was observed between the oldest adults (group A6: 76–99 years, mean 84 years) and the neonatal infants (group A1: r2 months, mean 26 days) (Table 2, see also Fig. 1). 3.2. Changes of GSH from birth to “teenage” years As shown in Table 2, in examination of the period of postnatal brain development and maturation, we found decreases or trends for a decrease of GSH levels with age during the first year of infancy in frontal cortex (r ¼ 0.52, n¼ 21, p ¼0.015), caudate (r ¼  0.47, n ¼22, p ¼ 0.027) and occipital cortex (r ¼  0.37, n ¼20, p ¼0.10) but not in the cerebellar cortex (r ¼  0.06, n ¼21, p ¼0.81) (see also Fig. 1, left panels). No changes in GSH levels were observed by Pearson correlation analysis from 1 to 18 years in any of the brain regions examined (Fig. 1, middle panels). Examination of the entire age range from birth (21 h) to 18 years of age disclosed declining GSH levels which were significant in frontal cortex (r ¼  0.50, n ¼33, p ¼0.003) and caudate (r ¼ 0.35, n ¼35, p ¼0.041). ANOVA of the three youngest age groups (A1–A3) also revealed significantly lower levels of GSH in group A3 (1–18 years) as compared to neonatal infants (A1, r2 months old) in frontal cortex (  55%) and caudate (  36%) but not in occipital (  20%) or cerebellar cortex (þ9%). 3.3. Changes of GSH during adulthood During adulthood from the ages of 23–99 years, we found positive correlations with age in levels of GSH in frontal cortex (r ¼ 0.45, n ¼37, p ¼0.005), cerebellar cortex (r ¼0.47, n ¼34, p ¼0.005) and caudate (r ¼0.34, n ¼35, p¼ 0.043) but not in occipital cortex (r ¼0.14, n ¼37, p ¼0.41) (Table 2; see also Fig. 1, right panels). This was mainly explained by difference between the elderly (450 years; A5 and A6) versus the younger adult (A4, 23–50 years) groups as no significant age-related change in GSH levels was observed from 23 to 50 years. ANOVA in the adult groups (A4– A6) showed significant differences (increases) in GSH levels between oldest (A6: 76–99 years, mean 84 years) and younger adults (A4: 23–50 years, mean 37 years) in frontal cortex (þ117%) but not

Fig. 1. Correlations (Pearson) between age and levels of reduced glutathione (GSH) in different brain regions examined during postnatal human brain development (first year infancy and from 1 to 18 years) and during adulthood (23–99 years). Only significant or trends for significant correlations (pr 0.10) are labeled.

in caudate (þ45%), cerebellar cortex (þ31%), or occipital cortex (þ13%). 3.4. Changes of GSSG during brain development and aging As expected [11,12,28], levels of GSSG were much lower and also much more variable than those of GSH, with average GSSG molar percentage of total glutathione [2  GSSG/(GSH þ2  GSSG)] ranging from 3.1 70.4% in cerebellar cortex to 8.6 7 0.9% in frontal cortex. Over the entire age range (21 days to 99 years), significant correlations between age and levels of GSSG were limited to frontal cortex (n ¼70; r ¼  0.35, p¼ 0.003), with GSSG levels in

114

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

adult groups (A4–A6) significantly lower than those in neonatal infants A1 (Table 3). In infants under one year of age, significant or trends of age-related decrease in levels of GSSG were observed in occipital cortex (r ¼ 0.75, n ¼20, p o0.0001), frontal cortex (r ¼  0.51, n¼ 21, p ¼0.018) and caudate (r ¼  0.35, n ¼22, p ¼0.11) but not in cerebellar cortex (r ¼ 0.06, n ¼20, p ¼ 0.79) (see Supplementary Fig. 1 for the plots). No further age-related changes in levels of GSSG were observed from 1 to 18 years or from 23 to 99 years with the exception of a significant positive correlation between age and GSSG levels in adults in caudate (23–99 years; r ¼0.44, n ¼36, p ¼0.007), which was explained by a marked (98%) increase in GSSG levels from young adult group A4 (23–50 years, mean 37 years) to the oldest adults (A6: 76–99 years, mean 84 years) (Table 3). No significant age-related change was observed for the molar ratio of GSH/GSSG in either the full age range, during postnatal development, or during adult brain aging with the exception of a positive correlation in the occipital cortex in neonates under one year of age (r ¼0.64, n ¼20, p ¼0.003) and in the full age range (r ¼0.30, n¼ 68, p ¼0.014) (see Supplementary Fig. 1 for the plots). 3.5. Potential confounding factors Negative slight correlations were observed between PMI and levels of GSH in occipital cortex (r ¼  0.35, n¼ 69, p¼ 0.003; see Supplementary Fig. 2A), frontal cortex (r ¼  0.27, n ¼70, p ¼0.022), and caudate (r ¼–0.23, n ¼70, p¼ 0.057) and between PMI and levels of GSSG in occipital cortex (r ¼ 0.37, n ¼68, p ¼0.002; Supplementary Fig. 2A), frontal cortex (r ¼  0.30, n¼ 70, p ¼0.012), and caudate (r ¼  0.37, n ¼71, p ¼0.001). There was no significant correlation between PMI and levels of GSH or GSSG in the cerebellar cortex. Controlling for PMI in partial correlations of age and levels of GSH or GSSG did not alter the significance or lack of significance of the age-glutathione correlations described above as there was no significant correlation between PMI and age of the subjects overall or in separate age groups (Supplementary Fig. 2B). Age related changes in protein concentrations in brain might have influenced the results of our study. In this regard, our data on levels of GSH and GSSG were expressed per protein content. In our investigation, we found that tissue protein content (per tissue wet weight) was generally stable during adult aging and during later (1–18 years) postnatal brain development. However, as expected [29,30], brain protein content per tissue wet weight was low at birth and increased linearly until approximately one year of age (Fig. 2A). Adjustment of the glutathione values to take into account this consideration, i.e., using levels of GSH and GSSG expressed per tissue wet weight, decreased the values of GSH and GSSG of the infants (A1 and A2) relative to other groups and in so doing eliminated the negative correlations between GSH or GSSG levels and age from 21 h to 1 year or later as described above. However, this correction did not alter the key findings during adult aging. Indeed, levels of another protein NSE, often used as a housekeeping “control” protein in biochemical studies, did not change with adult aging, in contrast to the age-related increase in GSH (see Fig. 2B). It is possible that the medical state of the subjects just prior to death might somehow have influenced the data. Attempts were made to categorize the subjects into four agonal subgroups according to the criteria of Hardy [24] although most of the subjects included in this study died of cardiovascular illnesses (Table 1). No significant effect of agonal status (ANCOVA with age as a covariate) was observed on levels of GSH or GSSG in brain regions examined and controlling for the cause of death (partial correlations) did not alter the significance or lack of significance of any of the age-related correlations described above (data not shown). Similarly, no significant effect of gender or brain laterality when controlling for

Fig. 2. (A) Changes in protein content per tissue wet weight in caudate nucleus during human brain development and aging. The solid line shows fitting with the hyperbolic function. Note the linear increase in protein/tissue ratio during first year infancy (dashed line, 0–1 year; Pearson r ¼0.62, p¼ 0.002). Afterwards, the ratio was stable at 0.054 70.001 (mean7 SEM) throughout the life. (B) Comparison of aging changes in levels of reduced glutathione (GSH) and neuron specific enolase (NSE; data were previously published in Tong et al. 2013 [27]) in the prefrontal cortex. Levels were normalized against respective means. F-test suggested that the slopes of linear regression with age were significantly different between GSH and NSE.

age (ANCOVAs) was observed on levels of GSH or GSSG in any of the brain regions. Analyzes of male and female subjects separately showed regression coefficients generally similar to those of the combined subjects (see Supplementary Fig. 3 for the plots).

4. Discussion The main finding of our study is that, in contradistinction to some preclinical animal data, brain levels of GSH did not decline with advancing age from early adulthood to senescence in the human. This suggests that aging might not compromise the glutathione antioxidant system in the adult human brain. 4.1. Methodological considerations and study limitations A variety of analytical procedures have been developed for measurement of tissue levels of GSH and GSSG, including colorimetric, fluorimetric, enzymatic and chromatographic assays with or without derivatization steps (see [31] for a review). These procedures differ with respect to complexity, sensitivity, specificity and reproducibility. The HPLC-coulometric electrochemical

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

detection procedure we employed is widely regarded as a preferred method, despite the instrumental requirements, for its many advantages including no sample manipulation or derivatization, high specificity, superior baseline stability, high efficiency, and simultaneous detection of both GSH and GSSG [32,33]. Autopsied brain levels of GSH and GSSG obtained in our assays were consistent with values obtained by other groups using similar or different detection methods [9–11,28,34,35]. Further, in a positive “neurological control” investigation [12], we detected, using our assay procedure, the expected loss of GSH in postmortem substantia nigra of patients with Parkinson’s disease [9–11]. Nevertheless, given the low abundance of GSSG as compared to GSH and its (expected) variability [10,11,19,28,35–37], interpretation of the data on GSSG concentrations and the ratios of GSH/GSSG should be made cautiously. A generic limitation of our study is the possibility that postmortem breakdown of GSH and GSSG [38] might have somehow skewed our results, particularly in view of our previous report of markedly lower levels of GSH and GSSG in autopsied versus biopsied human brain [12] and by our observation of significant negative correlations between postmortem interval and levels of GSH or GSSG in several brain regions. However, the lack of correlation between postmortem time and the age of our subjects and the absence of influence of postmortem time on the age-related correlations with the outcome measures suggest that our findings, in particular our inability to find age-related decrease in GSH levels during adulthood, are unlikely to be explained by the postmortem confound. Other limitations include uncertainty in agonal status of the subjects immediately preceding death and little information available on medications used by the subjects, in particular in the aged group immediately before death. For example, it is possible that our observation of elevated GSH levels with aging in several brain regions, or the absence of a GSH reduction could in part be related to increased medication (e.g., analgesics [39], blood pressure medications [40], sleeping pills [41]) in aged versus young subjects that might somehow modulate the GSH system in brain. Our sample was also imbalanced in gender distribution, particularly in younger adults with males dominating the 18–50 age range. Therefore, the lack of gender difference in levels and aging changes of GSH should be interpreted with caution. Most of the subjects included in this study (61%) died of cardiovascular illnesses. It can be argued that our findings might apply only to this particular group of people instead of general human “healthy aging”. 4.2. Changes in levels of GSH during human brain aging: our postmortem brain findings versus literature data It is difficult to compare our postmortem human brain data with the prior literature on the subject as the animal data, as a whole, are generally contradictory and the human findings are scanty. For example, in studies of the rat, during the first few months of life brain GSH levels were reported to decrease [36,42– 44], remain stable [45,46], or increase [47,48] (see also [49,50]). Nevertheless, our observation of different rate of regional GSH changes from infancy through adolescence to adulthood might possibly involve many processes of brain development and maturation including neurogenesis, gliogenesis, apoptosis, myelination, synaptogenesis and synaptic pruning (e.g., see [51]). For example, decreasing levels of GSH in the forebrain during postnatal brain development might be associated with an important role of the antioxidant and redox signaling in the switch from cell proliferation to differentiation [52]. Since GSH appears to be slightly predominant in glia versus neurons [48], our observation of increasing levels of GSH with adult aging in some brain regions

115

could be associated with increased gliosis with human brain aging [53]. The “GSH deficiency” hypothesis of brain aging has been supported by many experimental reports showing decreased levels of GSH in brain of aged versus young or mature animals including mice ([20,54]; but see [55,56]), rats ([18,19,36,57–59]; but see [56,60–62]), gerbils ([63]; but see [64]), dogs [65], fishes [66], and flies [67]. Negative findings were suggested to be possibly related to the aged not being examined at the true senescent stage [13] even though many of the conflicting reports used the same strain of animals at the same age range (e.g., [37] versus [55] in male C57BL/6 mice and [58] versus [61] in male Wistar rats). Recent studies suggest that changes in brain GSH levels during aging can be species, strain, gender, and regionally specific [68] – but if this statement is accurate, it would appear that an aging-related brain GSH decline in the animal literature is not a robust finding. Thus, in the senescence-resistant LOU rats, aging-loss of brain GSH was found only in female but not in male rats, with the striatum and hippocampus more vulnerable [68] although previous studies of other strains of rats (e.g., Sprague–Dawley) did not report gender differences [69]. A study of C57BL/6 mice showed instead that the males were more susceptible to aging-associated decline of brain GSH content than the females [70]. Another study of male Sprague–Dawley rats showed age-related decrease in GSH levels but limited to cerebral cortex (not in cerebellum or brainstem) [71]. In autopsied human brain, to our knowledge, the only study of GSH in aging that examined a similar lifespan as in the current study was the investigation of Venkateshappa and colleagues [22,23] which reported a negative correlation between age (0.01– 80 years) and total GSH levels in hippocampus. However, interpretation of the data is uncertain because of limited sample size (n¼ 14-22), especially the use of subjects who died of head trauma, and the generally low GSH values compared to those of ours and other studies – in particular, with close to zero values in the aged subjects that was driving the negative correlations [22]. A recent 1H MRS study reported a 35% loss of GSH in occipital lobe of elderly (77 years) as compared to young (20 years) subjects [21] – a finding not consistent with our observations in occipital cortex. The reason for this “discrepancy” is not clear but might be in part due to the non-ideal conditions, e.g., a 50% Cramer-Rao lower bound (CRLB), used for GSH detection in the 1H MRS study. 4.3. GSH, aging, and oxidative stress in human brain We found no evidence in support of the notion that levels of tissue antioxidants, in this case, GSH, decline with advancing age in adult human brain [13,21,72–74]. Instead, concentrations of GSH in the oldest adult brain (A6 age group, mean 84 years) were similar or even higher (in frontal cortex) as compared to those in developing brain (A3 age group, mean 8 years). Rather than a GSH decline, we found that GSH showed an increase with advancing age in several adult brain regions. This might possibly be explained by an age-related increase in oxidative stress in human brain. Thus, in some experimental animal studies, compensatory upregulation of GSH levels have been reported under a variety of conditions of mild to moderate oxidative stress, including environmental exposure to toxins such as mercury [75] and radiation [46], neonatal alcohol exposure [76], acute methamphetamine treatment [77], vitamin C deficit [78], and animal models of Parkinson’s disease (partial 6-hydroxydopamine lesion, [79]; young parkin null mice, [80]), Huntington’s disease (young R6/2 mice, [81]) and Alzheimer’s disease (apolipoprotein E null mice, [82]). The glutathione antioxidant response mechanism of the human brain might be sufficiently robust to meet mild to moderate oxidative stress challenges by elevating concentrations of GSH to maintain redox homeostasis. Perhaps only under extreme

116

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

conditions of oxidative stress and damage, for examples, in end stage Parkinson’s disease [12], brain GSH would be exhausted when recycling and/or de novo synthesis mechanisms are compromised and depletion of the antioxidant would further contribute to the deterioration of the conditions. Under the condition of human healthy aging, brain GSH concentrations may not be depleted but could even be compensatorily increased in response to external or internal challenges, e.g., chronic inflammatory reaction. It would be assumed that such a GSH increase would help protect the brain from oxidative stress. However, a recent study showed surprisingly that moderate overproduction of GSH in the midbrain of rats through overexpression of glutamate–cysteine ligase, as well as GSH depletion, leads to degeneration of dopaminergic neurons by inducing aberrant glutathiolation of cellular proteins [83].

5. Conclusions We found in autopsied human brain that levels of the major antioxidant GSH were not decreased during adult aging, a finding that might not be predicted based on a somewhat conflicting preclinical literature. Although our observations do suggest that the brain glutathione system is not compromised in the aging adult human, we consider our data to be preliminary and to suffer from the generic limitations inherent in all postmortem brain studies. Although measurement of GSH in living brain is not trivial, we suggest that a “replication” investigation might focus on measurement in aging human living brain using the still technically challenging and evolving MRS procedures.

Disclosure statement The authors have no conflicts of interest to disclose.

Acknowledgments This study was supported in part by the US NIDA/NIH DA07182 (SK), the New Zealand Institute of Environmental Science and Research, Ltd. (PF, SK), and the Centre for Addiction and Mental Health Foundation.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed. 2016.01.029.

References [1] D. Harman, Aging: a theory based on free radical and radiation chemistry, J. Gerontol. 11 (1956) 298–300. [2] D. Harman, The aging process, Proc. Natl. Acad. Sci. USA 78 (1981) 7124–7128. [3] B.N. Ames, M.K. Shigenaga, T.M. Hagen, Oxidants, antioxidants, and the degenerative diseases of aging, Proc. Natl. Acad. Sci. USA 90 (1993) 7915–7922. [4] R.A. Floyd, K. Hensley, Oxidative stress in brain aging. Implications for therapeutics of neurodegenerative diseases, Neurobiol. Aging 23 (2002) 795–807. [5] D. Fusco, G. Colloca, M.R. Lo Monaco, M. Cesari, Effects of antioxidant supplementation on the aging process, Clin. Interv. Aging 2 (2007) 377–387. [6] I. Sadowska-Bartosz, G. Bartosz, Effect of antioxidants supplementation on aging and longevity, Biomed. Res. Int. 2014 (2014) 404680. [7] R. Dringen, Metabolism and functions of glutathione in brain, Prog. Neurobiol. 62 (2000) 649–671. [8] T.L. Perry, D.V. Godin, S. Hansen, Parkinson’s disease: a disorder due to nigral glutathione deficiency? Neurosci. Lett. 33 (1982) 305–310.

[9] T.L. Perry, V.W. Yong, Idiopathic Parkinson’s disease, progressive supranuclear palsy and glutathione metabolism in the substantia nigra of patients, Neurosci. Lett. 67 (1986) 269–274. [10] E. Sofic, K.W. Lange, K. Jellinger, P. Riederer, Reduced and oxidized glutathione in the substantia nigra of patients with Parkinson’s disease, Neurosci. Lett. 142 (1992) 128–130. [11] J. Sian, D.T. Dexter, A.J. Lees, S. Daniel, Y. Agid, F. Javoy-Agid, P. Jenner, C. D. Marsden, Alterations in glutathione levels in Parkinson’s disease and other neurodegenerative disorders affecting basal ganglia, Ann. Neurol. 36 (1994) 348–355. [12] P.S. Fitzmaurice, L. Ang, M. Guttman, A.H. Rajput, Y. Furukawa, S.J. Kish, Nigral glutathione deficiency is not specific for idiopathic Parkinson’s disease, Mov. Disord. 18 (2003) 969–976. [13] P. Maher, The effects of stress and aging on glutathione metabolism, Ageing Res. Rev. 4 (2005) 288–314. [14] T. Cocco, P. Sgobbo, M. Clemente, B. Lopriore, I. Grattagliano, M. Di Paola, G. Villani, Tissue-specific changes of mitochondrial functions in aged rats: effect of a long-term dietary treatment with N-acetylcysteine, Free Radic. Biol. Med. 38 (2005) 796–805. [15] R.V. Sekhar, S.G. Patel, A.P. Guthikonda, M. Reid, A. Balasubramanyam, G. E. Taffet, F. Jahoor, Deficient synthesis of glutathione underlies oxidative stress in aging and can be corrected by dietary cysteine and glycine supplementation, Am. J. Clin. Nutr. 94 (2011) 847–853. [16] C.B. Pocernich, D.A. Butterfield, Elevation of glutathione as a therapeutic strategy in Alzheimer disease, Biochim. Biophys. Acta 1822 (2012) 625–630. [17] G.A. Hazelton, C.A. Lang, Glutathione contents of tissues in the aging mouse, Biochem. J. 188 (1980) 25–30. [18] M.Y. Farooqui, W.W. Day, D.M. Zamorano, Glutathione and lipid peroxidation in the aging rat, Comp. Biochem. Physiol. B 88 (1987) 177–180. [19] G. Benzi, O. Pastoris, F. Marzatico, R.F. Villa, Influence of aging and drug treatment on the cerebral glutathione system, Neurobiol. Aging 9 (1988) 371–375. [20] T.S. Chen, J.P. Richie Jr, C.A. Lang, The effect of aging on glutathione and cysteine levels in different regions of the mouse brain, Proc. Soc. Exp. Biol. Med. 190 (1989) 399–402. [21] U.E. Emir, D. Deelchand, P.G. Henry, M. Terpstra, Noninvasive quantification of T2 and concentrations of ascorbate and glutathione in the human brain from the same double-edited spectra, NMR Biomed. 24 (2011) 263–269. [22] C. Venkateshappa, G. Harish, A. Mahadevan, M.M. Srinivas Bharath, S. K. Shankar, Elevated oxidative stress and decreased antioxidant function in the human hippocampus and frontal cortex with increasing age: implications for neurodegeneration in Alzheimer’s disease, Neurochem. Res. 37 (2012) 1601–1614. [23] C. Venkateshappa, G. Harish, R.B. Mythri, A. Mahadevan, M.M. Bharath, S. K. Shankar, Increased oxidative damage and decreased antioxidant function in aging human substantia nigra compared to striatum: implications for Parkinson's disease, Neurochem. Res. 37 (2012) 358–369. [24] J.A. Hardy, P. Wester, B. Winblad, C. Gezelius, G. Bring, A. Eriksson, The patients dying after long terminal phase have acidotic brains; implications for biochemical measurements on autopsy tissue, J. Neural Transm. 61 (1985) 253–264. [25] S.J. Kish, K. Shannak, O. Hornykiewicz, Uneven pattern of dopamine loss in the striatum of patients with idiopathic Parkinson’s disease. Pathophysiologic and clinical implications, N. Engl. J. Med. 318 (1988) 876–880. [26] R.A. Riley, An atlas of the basal ganglia, brain stem and spinal cord, The Williams and Wilkins Company, Baltimore, MD, 1943. [27] J. Tong, J.H. Meyer, Y. Furukawa, I. Boileau, L.J. Chang, A.A. Wilson, S. Houle, S. J. Kish, Distribution of monoamine oxidase proteins in human brain: implications for brain imaging studies, J. Cereb. Blood Flow Metab. 33 (2013) 863–871. [28] A. Slivka, M.B. Spina, G. Cohen, Reduced and oxidized glutathione in human and monkey brain, Neurosci. Lett. 74 (1987) 112–118. [29] B.A. Holland, D.K. Haas, D. Norman, M. Brant-Zawadzki, T.H. Newton, MRI of normal brain maturation, Am. J. Neuroradiol. 7 (1986) 201–208. [30] P. Christiansen, P.B. Toft, P. Gideon, E.R. Danielsen, P. Ring, O. Henriksen, MRvisible water content in human brain: a proton MRS study, Magn. Reson. Imaging 12 (1994) 1237–1244. [31] P. Monostori, G. Wittmann, E. Karg, S. Turi, Determination of glutathione and glutathione disulfide in biological samples: an in-depth review, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 877 (2009) 3331–3346. [32] P.R. Harvey, R.G. Ilson, S.M. Strasberg, The simultaneous determination of oxidized and reduced glutathiones in liver tissue by ion pairing reverse phase high performance liquid chromatography with a coulometric electrochemical detector, Clin. Chim. Acta 180 (1989) 203–212. [33] I. Rebrin, R.S. Sohal, Pro-oxidant shift in glutathione redox state during aging, Adv. Drug Deliv. Rev. 60 (2008) 1545–1552. [34] J.D. Adams Jr, L.K. Klaidman, I.N. Odunze, H.C. Shen, C.A. Miller, Alzheimer’s and Parkinson’s disease. Brain levels of glutathione, glutathione disulfide, and vitamin E, Mol. Chem. Neuropathol. 14 (1991) 213–226. [35] S. Rose, S. Melnyk, O. Pavliv, S. Bai, T.G. Nick, R.E. Frye, S.J. James, Evidence of oxidative damage and inflammation associated with low glutathione redox status in the autism brain, Transl. Psychiatry 2 (2012) e134. [36] T. Iantomasi, F. Favilli, P. Marraccini, M. Stio, C. Treves, A. Quattrone, S. Capaccioli, M.T. Vincenzini, A. Quatrone, Age and GSH metabolism in rat cerebral cortex, as related to oxidative and energy parameters, Mech. Ageing Dev. 70 (1993) 65–82.

J. Tong et al. / Free Radical Biology and Medicine 93 (2016) 110–117

[37] I. Rebrin, S. Kamzalov, R.S. Sohal, Effects of age and caloric restriction on glutathione redox state in mice, Free Radic. Biol. Med. 35 (2003) 626–635. [38] T.L. Perry, S. Hansen, S.S. Gandham, Postmortem changes of amino compounds in human and rat brain, J. Neurochem. 36 (1981) 406–410. [39] J.P. De La Cruz, A. Guerrero, J.A. Gonzalez-Correa, M.M. Arrebola, S.F. de la Cuesta, Antioxidant effect of acetylsalicylic and salicylic acid in rat brain slices subjected to hypoxia, J. Neurosci. Res. 75 (2004) 280–290. [40] E.M. de Cavanagh, F. Inserra, L. Ferder, C.G. Fraga, Enalapril and captopril enhance glutathione-dependent antioxidant defenses in mouse tissues, Am. J. Physiol. Regul. Integr. Comp. Physiol. 278 (2000) R572–R577. [41] S.S. Hotta, Effect of chlorpromazine and amytal on the reduction of glutathione, Arch. Biochem. Biophys. 113 (1966) 395–398. [42] C.L. Kirstein, R. Coopersmith, R.J. Bridges, M. Leon, Glutathione levels in olfactory and non-olfactory neural structures of rats, Brain Res. 543 (1991) 341–346. [43] A. Sahoo, G.B. Chainy, Alterations in the activities of cerebral antioxidant enzymes of rat are related to aging, Int. J. Dev. Neurosci. 15 (1997) 939–948. [44] A.R. Rao, H. Quach, E. Smith, G.T. Vatassery, R. Rao, Changes in ascorbate, glutathione and alpha-tocopherol concentrations in the brain regions during normal development and moderate hypoglycemia in rats, Neurosci. Lett. 568 (2014) 67–71. [45] S.O. Rahaman, S. Ghosh, K.P. Mohanakumar, S. Das, P.K. Sarkar, Hypothyroidism in the developing rat brain is associated with marked oxidative stress and aberrant intraneuronal accumulation of neurofilaments, Neurosci. Res. 40 (2001) 273–279. [46] C.G. Di Toro, P.A. Di Toro, L.M. Zieher, L.R. Guelman, Sensitivity of cerebellar glutathione system to neonatal ionizing radiation exposure, Neurotoxicology 28 (2007) 555–561. [47] H. Kudo, T. Kokunai, T. Kondoh, N. Tamaki, S. Matsumoto, Quantitative analysis of glutathione in rat central nervous system: comparison of GSH in infant brain with that in adult brain, Brain Res. 511 (1990) 326–328. [48] M.E. Rice, I. Russo-Menna, Differential compartmentalization of brain ascorbate and glutathione between neurons and glia, Neuroscience 82 (1998) 1213–1223. [49] S. Berl, D.P. Purpura, Postnatal changes in amino acid content of kitten cerebral cortex, J. Neurochem. 10 (1963) 237–240. [50] D. Nanda, J. Tolputt, K.J. Collard, Changes in brain glutathione levels during postnatal development in the rat, Brain Res. Dev. Brain Res. 94 (1996) 238–241. [51] P.R. Huttenlocher, A.S. Dabholkar, Regional differences in synaptogenesis in human cerebral cortex, J. Comp. Neurol. 387 (1997) 167–178. [52] E.A. Ostrakhovitch, O.A. Semenikhin, The role of redox environment in neurogenic development, Arch. Biochem. Biophys. 534 (2013) 44–54. [53] J. Tong, Y. Furukawa, A. Sherwin, O. Hornykiewicz, S.J. Kish, Heterogeneous intrastriatal pattern of proteins regulating axon growth in normal adult human brain, Neurobiol. Dis. 41 (2011) 458–468. [54] T. Sasaki, M. Senda, S. Kim, S. Kojima, A. Kubodera, Age-related changes of glutathione content, glucose transport and metabolism, and mitochondrial electron transfer function in mouse brain, Nucl. Med. Biol. 28 (2001) 25–31. [55] S. Hussain, W. Slikker Jr, S.F. Ali, Age-related changes in antioxidant enzymes, superoxide dismutase, catalase, glutathione peroxidase and glutathione in different regions of mouse brain, Int. J. Dev. Neurosci. 13 (1995) 811–817. [56] J.G. de la Asuncion, A. Millan, R. Pla, L. Bruseghini, A. Esteras, F.V. Pallardo, J. Sastre, J. Vina, Mitochondrial glutathione oxidation correlates with age-associated oxidative damage to mitochondrial DNA, FASEB J. 10 (1996) 333–338. [57] V. Ravindranath, B.R. Shivakumar, H.K. Anandatheerthavarada, Low glutathione levels in brain regions of aged rats, Neurosci. Lett. 101 (1989) 187–190. [58] S.K. Sandhu, G. Kaur, Alterations in oxidative stress scavenger system in aging rat brain and lymphocytes, Biogerontology 3 (2002) 161–173. [59] V. Calabrese, G. Scapagnini, A. Ravagna, C. Colombrita, F. Spadaro, D. A. Butterfield, A.M. Giuffrida Stella, Increased expression of heat shock proteins in rat brain during aging: relationship with mitochondrial function and glutathione redox state, Mech. Ageing Dev. 125 (2004) 325–335. [60] L.E. Rikans, D.R. Moore, Effect of aging on aqueous-phase antioxidants in tissues of male Fischer rats, Biochim. Biophys. Acta 966 (1988) 269–275.

117

[61] G. Barja de Quiroga, R. Perez-Campo, M. Lopez Torres, Anti-oxidant defences and peroxidation in liver and brain of aged rats, Biochem. J. 272 (1990) 247–250. [62] J.L. Harris, H.W. Yeh, R.H. Swerdlow, I.Y. Choi, P. Lee, W.M. Brooks, High-field proton magnetic resonance spectroscopy reveals metabolic effects of normal brain aging, Neurobiol. Aging 35 (2014) 1686–1694. [63] R.S. Sohal, S. Agarwal, B.H. Sohal, Oxidative stress and aging in the Mongolian gerbil (Meriones unguiculatus), Mech. Ageing Dev. 81 (1995) 15–25. [64] J.R. Zhang, P.K. Andrus, E.D. Hall, Age-related regional changes in hydroxyl radical stress and antioxidants in gerbil brain, J. Neurochem. 61 (1993) 1640–1647. [65] E. Head, J. Liu, T.M. Hagen, B.A. Muggenburg, N.W. Milgram, B.N. Ames, C. W. Cotman, Oxidative damage increases with age in a canine model of human brain aging, J. Neurochem. 82 (2002) 375–381. [66] M.E. Gotz, C.R. Malz, A. Dirr, D. Blum, W. Gsell, S. Schmidt, R. Burger, S. Pohli, P. Riederer, Brain aging phenomena in migrating sockeye salmon Oncorhynchus nerka nerka, J. Neural. Transm. (Vienna) 112 (2005) 1177–1199. [67] M. Haddadi, S.R. Jahromi, B.K. Sagar, R.K. Patil, T. Shivanandappa, S.R. Ramesh, Brain aging, memory impairment and oxidative stress: a study in Drosophila melanogaster, Behav. Brain Res. 259 (2014) 60–69. [68] E. Moyse, M. Arseneault, P. Gaudreau, G. Ferland, C. Ramassamy, Gender- and region-dependent changes of redox biomarkers in the brain of successfully aging LOU/C rats, Mech. Ageing. Dev. 149 (2015) 19–30. [69] Y. Zhu, P.M. Carvey, Z. Ling, Age-related changes in glutathione and glutathione-related enzymes in rat brain, Brain Res. 1090 (2006) 35–44. [70] H. Wang, H. Liu, R.M. Liu, Gender difference in glutathione metabolism during aging in mice, Exp. Gerontol. 38 (2003) 507–517. [71] M. Benuck, M. Banay-Schwartz, T. DeGuzman, A. Lajtha, Effect of food deprivation on glutathione and amino acid levels in brain and liver of young and aged rats, Brain Res. 678 (1995) 259–264. [72] A. Currais, P. Maher, Functional consequences of age-dependent changes in glutathione status in the brain, Antioxid. Redox Signal 19 (2013) 813–822. [73] S. Saharan, P.K. Mandal, The emerging role of glutathione in Alzheimer’s disease, J. Alzheimers Dis. 40 (2014) 519–529. [74] I.Y. Choi, P. Lee, D.R. Denney, K. Spaeth, O. Nast, L. Ptomey, A.K. Roth, J. A. Lierman, D.K. Sullivan, Dairy intake is associated with brain glutathione concentration in older adults, Am. J. Clin. Nutr. 101 (2015) 287–293. [75] D.J. Hoffman, M.G. Spalding, P.C. Frederick, Subchronic effects of methylmercury on plasma and organ biochemistries in great egret nestlings, Environ. Toxicol. Chem. 24 (2005) 3078–3084. [76] A.M. Smith, D.R. Zeve, J.J. Grisel, W.J. Chen, Neonatal alcohol exposure increases malondialdehyde (MDA) and glutathione (GSH) levels in the developing cerebellum, Brain Res. Dev. Brain Res. 160 (2005) 231–238. [77] C. Harold, T. Wallace, R. Friedman, G. Gudelsky, B. Yamamoto, Methamphetamine selectively alters brain glutathione, Eur. J. Pharmacol. 400 (2000) 99–102. [78] F.E. Harrison, M.E. Meredith, S.M. Dawes, J.L. Saskowski, J.M. May, Low ascorbic acid and increased oxidative stress in gulo(-/-) mice during development, Brain Res. 1349 (2010) 143–152. [79] Y. Aluf, J. Vaya, S. Khatib, Y. Loboda, S. Kizhner, J.P. Finberg, Specific oxidative stress profile associated with partial striatal dopaminergic depletion by 6-hydroxydopamine as assessed by a novel multifunctional marker molecule, Free Radic. Res. 44 (2010) 635–644. [80] J.A. Rodriguez-Navarro, M.J. Casarejos, J. Menendez, R.M. Solano, I. Rodal, A. Gomez, J.G. Yebenes, M.A. Mena, Mortality, oxidative stress and tau accumulation during ageing in parkin null mice, J. Neurochem. 103 (2007) 98–114. [81] I. Tkac, J.M. Dubinsky, C.D. Keene, R. Gruetter, W.C. Low, Neurochemical changes in Huntington R6/2 mouse striatum detected by in vivo 1H NMR spectroscopy, J. Neurochem. 100 (2007) 1397–1406. [82] F. Tchantchou, M. Graves, E. Rogers, D. Ortiz, T.B. Shea, N-acteyl cysteine alleviates oxidative damage to central nervous system of ApoE-deficient mice following folate and vitamin E-deficiency, J. Alzheimers Dis. 7 (2) (2005) 135–138 (discussion 173-80). [83] M. Garrido, Y. Tereshchenko, Z. Zhevtsova, G. Taschenberger, M. Bahr, S. Kugler, Glutathione depletion and overproduction both initiate degeneration of nigral dopaminergic neurons, Acta Neuropathol. 121 (2011) 475–485.