Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jun 26.
Published in final edited form as: Drug Alcohol Depend. 2025 May 12;272:112705. doi: 10.1016/j.drugalcdep.2025.112705

Brain glutathione levels and associations with recent drinking in treatment-naïve individuals with alcohol use disorder versus light drinkers

James J Prisciandaro a,*, Joseph P Schacht b, Andrew P Prescot c, Raymond F Anton a
PMCID: PMC13295321  NIHMSID: NIHMS2176685  PMID: 40359607

Abstract

Background and aims:

Repeated ethanol exposure produces excess oxidative stress resulting in cellular damage, with glutathione (GSH), the brain’s primary antioxidant, conferring a critical first line of defense. This study aimed to compare brain GSH levels between treatment-naïve individuals with Alcohol Use Disorder (AUD) and light drinking control participants (LD). This study also aimed to evaluate associations of brain GSH levels with recent heavy drinking in AUD and LD participants.

Design setting and participants:

Secondary analyses were conducted on cross-sectional neuroimaging data from (n = 20) treatment-naïve individuals with AUD and (n = 20) demographically matched LD participants at a medical university in South Carolina, USA.

Measurements:

CSF-corrected and water-referenced GSH levels from dorsal anterior cingulate cortex (dACC) acquired via proton MR spectroscopy and past 14-day number of heavy drinking days (nHDD) acquired via Time-Line Followback interview.

Findings:

Significantly higher dACC GSH/water levels were observed in AUD participants (M = 0.88, SD = 0.16) than in LD participants (M = 0.70, SD = 0.16, F = 12.00, p < 0.001; Cohen’s d = 1.10). Additionally, there was a significant interaction between GSH/water levels and participant group (F = 5.71, p = 0.022), such that higher GSH/water levels were associated with lower nHDD in AUD (r = −0.46, p = 0.040) but not LD (r = 0.24 p = 0.329) participants.

Conclusions:

The findings from this preliminary study are consistent with an interpretation of compensatory GSH upregulation in response to moderate oxidative stress in treatment-naïve individuals with AUD, adding unique support to oxidative stress models of alcohol-related cellular damage and highlighting the potential promise of antioxidant treatments for AUD.

Keywords: Alcohol use disorder, Proton MR spectroscopy, Glutathione, GSH, Oxidative stress, 1H-MRS, Treatment-naive

1. Introduction

Repeated ethanol (“alcohol”) exposure produces excess oxidative stress resulting in cellular damage, with glutathione (GSH), the brain’s primary antioxidant, conferring a critical first line of defense against this phenomenon. Alcohol metabolism generates substantial toxic free radicals (Haorah et al., 2008), with its metabolic product, acetaldehyde, consuming free GSH (Rodriguez and Coveñas, 2021). Chronic alcohol consumption (“drinking”) indirectly produces additional oxidative stress, for example, through excess activation of glutamate receptors, particularly during early alcohol withdrawal (Gass and Olive, 2008). In compensatory response to this elevated oxidative stress, GSH is upregulated (Duffy et al., 2014), neutralizing free radicals and reactive oxidants and providing protection and redox regulation of cellular thiol proteins via S-glutathionylation, a post-translational modification on cysteine residues (Townsend, 2007). With continued heavy drinking, and repeated bouts of alcohol withdrawal-mediated excitotoxicity, however, these defenses are eventually overwhelmed (Tsermpini et al., 2022), resulting in diminished peripheral GSH levels (Yang et al., 2022), along with oxidative damage to DNA, RNA, proteins, and lipids (Heymann et al., 2018). Despite the potential clinical implications of these findings, and the availability of imaging methods capable of measuring brain GSH levels in vivo (i.e., proton MR spectroscopy; 1H-MRS), we are unaware of any studies comparing brain GSH levels in vivo in people with AUD relative to healthy control participants. The present study addresses this gap in the literature by comparing brain GSH levels in treatment-naïve individuals with AUD versus light drinking participants, along with preliminary evaluation of associations between GSH levels and recent heavy drinking.

2. Methods

The present manuscript reports secondary analyses from the study detailed fully in (Prisciandaro et al., 2019) (available via PubMed Central [PMC], PMCID: PMC6538297). Briefly, twenty treatment-naïve individuals aged 21–40 years with AUD and 20 demographically matched light drinkers (LD) completed a Time-Line Followback (TLFB) daily drinking interview (Sobell and Sobell, 1995) for the previous 14 days followed by an MRI, featuring a T1-weighted structural scan for voxel placement and tissue segmentation and a 1H-MRS (2-dimensional J-resolved point-resolved spectroscopy; 2D J-PRESS; (Prescot and Renshaw, 2013)) scan of an 18.75 cm3 volume located in dorsal anterior cingulate cortex (dACC) on a 3 T SIEMENS TIM Trio scanner with a 32-channel headcoil (Erlangen, Germany); see 6 for representative voxel placement and GSH-labeled spectrum. Individuals taking psychoactive medications were excluded. We previously (Prisciandaro et al., 2019, 2016) reported that CSF-corrected and water-referenced dACC glutamine and GABA, but not glutamate, levels were reduced among AUD versus LD participants, while lower glutamate, but not glutamine or GABA levels, were associated with greater past 14-day number of heavy drinking days (nHDD). For this report, CSF-corrected and water-referenced dACC GSH levels were compared between AUD and LD participants and associations of dACC GSH levels with past 14-day nHDD, along with determination of whether such associations significantly differed between participant groups via the interaction of participant group with mean-centered GSH levels, were estimated using general linear modeling (GLM). In the case of a significant interaction of participant group with GSH levels in this second GLM model, within-group Pearson correlation coefficients were additionally estimated. Covariates evaluated in prior publications (i.e., education and gray matter to total brain tissue matter ratio [GM:BM]) and suggested by reviewers (age and sex) were considered here as well, but were ultimately removed from final models because they were neither significantly associated with dACC GSH levels or nHDD (ps > 0.5) nor did their inclusion substantively alter model results (i.e., the statistical significance of participant group, GSH levels, or their interaction did not change). Given the targeted nature of the present analyses (i.e., 2 statistical models, each having a different dependent variable), alpha was nominally set to 0.05 across analyses. We hypothesized that AUD and LD participants would have significantly different dACC GSH levels and that GSH would be significantly associated with nHDD in AUD but not LD participants.

3. Results

Full participant details are available in (Prisciandaro et al., 2019); most participants were men of European descent in their twenties who last drank within 7-days of MRI (AUD, M(SD)= 2.45(1.23); LD, M(SD)= 5.20(7.62); p < 0.001). AUD participants drank significantly more than LD participants (nHDD: AUD, M(SD)= 4.75(3.26); LD, M(SD)= 0.47 (0.77); p < 0.001; drinks/drinking day: AUD, M(SD)= 7.22(3.43); LD, M(SD)= 2.42(1.70); p < 0.001). GSH data were of excellent quality, with uniformly low Cramer-Rao Lower Bound (%) estimates (M=2.56, SD=0.98). Significantly higher dACC GSH/water levels were observed in AUD participants (M=0.88, SD=0.16) than in LD participants (M=0.70, SD=0.16, F=12.00, p < 0.001; Cohen’s d=1.10; Fig. 1). Additionally, there was a significant interaction between GSH/water levels and participant group (F=5.71, p = 0.022), such that higher GSH/water levels were associated with lower nHDD in AUD (r = −0.46, p = 0.040) but not LD (r = 0.24, p = 0.329) participants (Fig. 2). Given apparent non-linearity in the scatterplot of nHDD and GSH in AUD participants (6), we evaluated the relative fit of quadratic vs. linear lines and found the former provided better fit to the data (i.e., r2=.416 vs. r2=.215).

Fig. 1.

Fig. 1.

Open in a new tab

Dorsal anterior cingulate cortex glutathione (GSH) / water levels, CSF-corrected and expressed in Institutional Units, in treatment-naïve individuals with alcohol use disorder (AUD) versus light drinking (LD) participants.

Fig. 2.

Fig. 2.

Open in a new tab

Scatterplot and linear fit lines of associations of dorsal anterior cingulate cortex glutathione (GSH) / water levels, CSF-corrected and expressed in Institutional Units, with number of heavy drinking days within the past 14 days in treatment-naïve individuals with alcohol use disorder (AUD, in blue; r = −0.46, p = 0.040) and light drinking (LD, in red; r = 0.24, p = 0.329) participants.

4. Discussion

The present study is to our knowledge the first to compare brain GSH levels in vivo between people with AUD and healthy control (i.e., LD) participants, finding that treatment-naïve individuals with AUD had significantly higher dACC GSH levels than did LD participants. Highlighting the functional significance of this finding, higher brain GSH levels were significantly associated with fewer past 14-day heavy drinking days in AUD but not LD participants. These findings have potential implications for AUD treatment development.

Though preclinical (Waly et al., 2011) and postmortem (Götz et al., 2001) studies have reported diminished brain GSH levels with AUD, both increased and decreased brain GSH levels have been associated with moderate (e.g., mild cognitive impairment; (Duffy et al., 2014) and extreme (e.g., dementia; (Mazzetti et al., 2015) oxidative stress conditions, respectively. Accordingly, we cautiously preferred non-directional hypotheses. In the context of research on brain GSH levels among individuals with neurological disease, the finding that treatment-naïve individuals with AUD had relatively elevated levels of GSH is consistent with an interpretation of compensatory GSH upregulation in response to moderate oxidative stress. The finding that elevated GSH levels were associated with relatively fewer recent heavy drinking days, though seemingly counterintuitive, could indicate that AUD participants with relatively more recent heavy drinking days were experiencing greater-than-moderate levels of oxidative stress, resulting in declining GSH levels. Future research should evaluate relative brain GSH levels and associations with heavy drinking in samples including treatment-seeking AUD participants, who have more severe AUD and who likely experience more alcohol-related oxidative stress, as the relative direction of brain GSH levels could be different (i.e., significantly lower than LD) in such individuals. Also, sequentially scanning individuals during heavy drinking and abstinence might elucidate the dynamics of this alcohol consumption and withdrawal.

Although preliminary, the findings from this study may have implications for AUD treatment development. N-acetylcysteine (NAC), which increases GSH levels by providing cells with the rate-limiting component of GSH synthesis, cysteine, has been researched as a potential treatment for substance use disorders (McClure et al., 2014). Preclinical studies have demonstrated that NAC reduces alcohol drinking, motivation, seeking, and relapse in dependent animals (Lebourgeois et al., 2019, 2018). Early NAC RCTs showed decreased frequency and quantity of drinking in people with Cannabis Use Disorder (Squeglia et al., 2018), and decreased craving in people with co-occurring posttraumatic stress disorder and SUD, > 80 % of whom had AUD (Back et al., 2016). More recently, the first published RCT of NAC for AUD found that, relative to placebo, 28-days of NAC treatment decreased drinks per drinking day, though not number of heavy drinking days (Morley et al., 2023). Despite these potentially promising findings, no studies we are aware of have evaluated whether NAC-related reductions in drinking are associated with changes in brain GSH levels. Such studies could pave the way for the development of alternative and potentially more powerful antioxidant treatments for AUD.

Although the present study had several strengths, including a demographically matched LD control group and use of a 1H-MRS sequence capable of separating GSH from overlapping signals at 3 T, its limitations must also be acknowledged. First, although sample size was typical of 1H-MRS studies, these results should be considered tentative until replicated in larger samples. For example, the obtained superior quadratic vs. linear fit of the association of GSH with number of heavy drinking days was driven by 3 data points. Second, because we chose to compare AUD with LD participants, it is unknown the degree to which our results generalize to non-drinker (“teetotaler”) participants. Third, 1H-MRS data were acquired from a single brain region (i.e., dACC), constraining our ability to generalize findings from the present study to other brain regions. Fourth, while the 1H-MRS sequence used in the present study was a relative strength (i.e., compared to more commonly used sequences, like PRESS, which are unable to reliably distinguish GSH from overlapping signals at 3 T; c.f., (Moss et al., 2018), STEAM in neonates), a J-difference 1H-MRS sequence (e.g., MEGA-PRESS) or higher magnetic field strength may have resolved GSH even more successfully (Sanaei Nezhad et al., 2017). In conclusion, the results from the present study demonstrate for the first time that treatment naïve individuals with AUD have significantly greater brain GSH levels than LD, adding unique support to oxidative stress models of alcohol-related cellular damage and suggesting the promise of antioxidant treatments for AUD.

Supplementary Material

MMC2
MMC1

Funding

This study was funded by the National Institute on Alcohol Abuse and Alcoholism (P50 AA010761, PI: Becker).

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.drugalcdep.2025.112705.

Footnotes

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Conflict of interest disclosures

Drs. Prisciandaro and Schacht were funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Dr Anton reported grants from the NIAAA and a K05 Career Development Grant from the National Institutes of Health during the conduct of the study; personal fees from Alkermes, Denovo, Dicerna, Foxo Bioscience, Imbrium, Otsuka, and Saphosyne Pharma as consultant; and grants from Laboratorio Farmaceutico C.T. outside the submitted work. No other disclosures were reported.

CRediT authorship contribution statement

Anton Raymond F.: Writing – review & editing, Supervision, Conceptualization. Prisciandaro James J.: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Prescot Andrew P.: Writing – review & editing, Software, Methodology, Formal analysis. Schacht Joseph P.: Writing – review & editing, Project administration, Funding acquisition, Data curation.

References

  1. Back SE, McCauley JL, Korte KJ, Gros DF, Leavitt V, Gray KM, et al. , 2016. A double-blind, randomized, controlled pilot trial of N-acetylcysteine in veterans with posttraumatic stress disorder and substance use disorders. J. Clin. Psychiatry 77 (11), e1439–e1446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Duffy SL, Lagopoulos J, Hickie IB, Diamond K, Graeber MB, Lewis SJ, et al. , 2014. Glutathione relates to neuropsychological functioning in mild cognitive impairment. Alzheimers Dement 10 (1), 67–75. [DOI] [PubMed] [Google Scholar]
  3. Gass JT, Olive MF, 2008. Glutamatergic substrates of drug addiction and alcoholism. Biochem. Pharmacol. 75 (1), 218–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Götz ME, Janetzky B, Pohli S, Gottschalk A, Gsell W, Tatschner T, et al. , 2001. Chronic alcohol consumption and cerebral indices of oxidative stress: is there a link? Alcohol.: Clin. Exp. Res. 25 (5), 717–725. [PubMed] [Google Scholar]
  5. Haorah J, Ramirez SH, Floreani N, Gorantla S, Morsey B, Persidsky Y, 2008. Mechanism of alcohol-induced oxidative stress and neuronal injury. Free Radic. Biol. Med 45 (11), 1542–1550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Heymann HM, Gardner AM, Gross ER, 2018. Aldehyde-induced DNA and protein adducts as biomarker tools for alcohol use disorder. Trends Mol. Med 24 (2), 144–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Lebourgeois S, Gonzalez-Marín MC, Antol J, Naassila M, Vilpoux C, 2019.´ Evaluation of N-acetylcysteine on ethanol self-administration in ethanol-dependent rats. Neuropharmacology 150, 112–120. [DOI] [PubMed] [Google Scholar]
  8. Lebourgeois S, González-Marín MC, Jeanblanc J, Naassila M, Vilpoux C, 2018. Effect of N-acetylcysteine on motivation, seeking and relapse to ethanol self-administration. Addict. Biol. 23 (2), 643–652. [DOI] [PubMed] [Google Scholar]
  9. Mazzetti AP, Fiorile MC, Primavera A, Lo Bello M, 2015. Glutathione transferases and neurodegenerative diseases. Neurochem. Int 82, 10–18. [DOI] [PubMed] [Google Scholar]
  10. McClure EA, Gipson CD, Malcolm RJ, Kalivas PW, Gray KM, 2014. Potential role of N-acetylcysteine in the management of substance use disorders. CNS Drugs 28 (2), 95–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Morley KC, Peruch S, Adams C, Towers E, Tremonti C, Watt J, et al. , 2023. N acetylcysteine in the treatment of alcohol use disorder: a randomized, double-blind, placebo-controlled trial. Alcohol Alcohol 58 (5), 553–560. [DOI] [PubMed] [Google Scholar]
  12. Moss HG, Brown TR, Wiest DB, Jenkins DD, 2018. N-Acetylcysteine rapidly replenishes central nervous system glutathione measured via magnetic resonance spectroscopy in human neonates with hypoxic-ischemic encephalopathy. J. Cereb. Blood Flow. Metab. 38 (6), 950–958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Prescot AP, Renshaw PF, 2013. Two-dimensional J-resolved proton MR spectroscopy and prior knowledge fitting (ProFit) in the frontal and parietal lobes of healthy volunteers: assessment of metabolite discrimination and general reproducibility. J. Magn. Reson Imaging 37 (3), 642–651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Prisciandaro JJ, Schacht JP, Prescot AP, Renshaw PF, Brown TR, Anton RF, 2016. Associations between recent heavy drinking and dorsal anterior cingulate N-acetylaspartate and glutamate concentrations in non-treatment-seeking individuals with alcohol dependence. Alcohol., Clin. Exp. Res. 40 (3), 491–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Prisciandaro JJ, Schacht JP, Prescot AP, Renshaw PF, Brown TR, Anton RF, 2019. Brain glutamate, GABA, and glutamine levels and associations with recent drinking in treatment-naïve individuals with alcohol use disorder versus light drinkers. Alcohol., Clin. Exp. Res. 43 (2), 221–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Rodriguez FD, Coveñas R, 2021. Biochemical mechanisms associating alcohol use disorders with cancers. Cancers 13 (14). [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Sanaei Nezhad F, Anton A, Parkes LM, Deakin B, Williams SR, 2017. Quantification of glutathione in the human brain by MR spectroscopy at 3 Tesla: Comparison of PRESS and MEGA-PRESS. Magn. Reson Med 78 (4), 1257–1266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Sobell LC, Sobell MB, 1995. Alcohol Timeline Followback Users’ Manual. Addiction Research Foundation, Toronto, Canada. [Google Scholar]
  19. Squeglia LM, Tomko RL, Baker NL, McClure EA, Book GA, Gray KM, 2018. The effect of N-acetylcysteine on alcohol use during a cannabis cessation trial. Drug Alcohol Depend. 185, 17–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Townsend DM, 2007. S-glutathionylation: indicator of cell stress and regulator of the unfolded protein response. Mol. Inter. 7 (6), 313–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Tsermpini EE, Plemenitaš Ilješ A, Dolžan V, 2022. Alcohol-induced oxidative stress and the role of antioxidants in alcohol use disorder: a systematic review. Antioxidants 11 (7), 1374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Waly MI, Kharbanda KK, Deth RC, 2011. Ethanol lowers glutathione in rat liver and brain and inhibits methionine synthase in a cobalamin-dependent manner. Alcohol., Clin. Exp. Res. 35 (2), 277–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Yang M, Zhou X, Tan X, Huang X, Yuan L, Zhang Z, et al. , 2022. The status of oxidative stress in patients with alcohol dependence: a meta-analysis. Antioxidants 11 (10). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

MMC2
NIHMS2176685-supplement-MMC2.jpg (718.2KB, jpg)
MMC1
NIHMS2176685-supplement-MMC1.jpg (1.3MB, jpg)

RESOURCES