Dual Stages of Alcohol-Related Cerebral White Matter Degeneration Reviewed: Early-Stage Stress/Neuroinflammation Versus Late-Stage Impaired Insulin/IGF Signaling Through Akt-mTOR–Review

Abstract

Long-term effects of alcohol-related brain damage (ARBD) include neurocognitive and neurobehavioral dysfunctions with neurodegeneration. White matter (WM) is notably targeted across the lifespan yet relatively little is known about the stages, mechanisms, and consequences of myelin and axonal loss. In alcohol-related liver disease, early pathology is reversible, but with chronic heavy alcohol exposures, disease progresses with degeneration, and ultimately organ failure. Similarly, WM ARBD also develops in two broad stages. The early stages of WM ARBD are likely mediated by vascular dysfunction with tissue swelling, oligodendrocyte dysfunction, myelin loss, neuroinflammation, and oxidative stress. The chronic progressive stage is linked to metabolic dysfunction related to impairments in insulin and insulin-like growth factor signaling through Akt-mechanistic target of rapamycin (mTOR) pathways that mediate oligodendrocyte survival and function, myelin homeostasis, and blood-brain-barrier (BBB) integrity. We hypothesize that early-stage WM ARBD may be largely reversible by abstinence and anti-oxidant/anti-inflammatory measures, whereas late-stage ARBD requires strategies to restore WM/oligodendrocyte metabolic function via insulin sensitizer, antioxidant, anti-inflammatory, and myelin homeostasis/normalization support. Multi-pronged, overlapping but distinct therapeutic strategies are needed to reduce the impact and long-term health consequences of chronic progressive WM ARBD.

Alcohol-Related Brain Degeneration—Spotlight on White Matter

The consequences of alcohol abuse and addiction are among the costliest healthcare problems in the world. In the USA, alcohol abuse is the third leading preventable cause of death (88,000/year) (Centers for Disease Control & Prevention, 2008; Mokdad et al., 2004; Serdula et al., 2004). In Australia, one-third of adults drink alcohol at levels above the Australian Alcohol Guidelines, placing them at significant risk for alcohol-related morbidity from liver disease, neurodegeneration, cardiac disease, and increased mortality (Fetherston & Calder, 2023). In 2022, alcohol-related deaths reached the highest level of the previous decade (Fetherston & Calder, 2023). In adults, alcohol use disorders cause alcohol-related brain damage (ARBD) characterized by neurobehavioral abnormalities and cognitive deficits including impairments in executive functions (Chanraud et al., 2007) which can progress to neurodegeneration with dementia and disability (Li, 2008; Schmidt et al., 2005). An important and consistent neuroanatomical substrate of ARBD is brain atrophy (Harper, 1982) with selective loss of white matter (WM) including both myelin and axons (de la Monte, 1988; Harper et al., 1990; Sutherland et al., 2014). WM atrophy and myelin degeneration are linked to cognitive impairment (de la Monte & Kril, 2014; Elofson et al., 2013; Jacobus et al., 2013; Schmidt et al., 2005). Neuroimaging and postmortem studies have shown that the severity of WM atrophy correlates with maximum daily and lifetime alcohol exposures (de la Monte & Kril, 2014; Harper et al., 2003; Sutherland et al., 2014).

ARBD’s prominent targeting of the corpus callosum, prefrontal, temporal, and cerebellar WM (de la Monte & Kril, 2014; Phillips et al., 1987) indicates that the distribution of injury is non-random. Neuroimaging studies have focused on corpus callosal atrophy (Chanraud et al., 2007; Pfefferbaum et al., 1996; Pfefferbaum et al., 2007; Estruch et al., 1997) because the attendant disruption of inter-hemispheric communications compromises the exchange of sensory, motor, and cognitive information. In addition, the relative structural simplicity of the corpus callosum enables convenient and straight-forward quantitative assessments of atrophy. Diffusion tensor imaging data suggest that corpus callosal atrophy correlates with altered WM micro-structural integrity (Schulte et al., 2005; Pfefferbaum et al., 2006). Experimental animal studies corroborate human data, showing that heavy alcohol consumption leading to cognitive impairment causes WM atrophy with demyelination, dysmyelination, and axonal degeneration (Papp-Peka et al., 2017).

Combined results stemming from human and experimental animal model studies suggest that alcohol’s neurotoxic effects in white matter begin with demyelination and progress to variable degrees of axonal degeneration. Neuroimaging and postmortem neuropathological studies showed that short-term, high-dose alcohol exposures leading to acute neurological disorders such as Marchiafava-Bignami disease, alcohol-related encephalopathy, or Wernicke-Korsakoff syndrome are associated with myelin swelling, demyelination, and myelinolysis (Bourgouin et al., 1995; Marjama et al., 1994; Ostertun et al., 1990; Pfefferbaum & Sullivan, 2002; Tuntiyatorn & Laothamatas, 2008). In addition, experimental rodent models of ethanol exposure revealed early, predominantly demyelinating pathology in WM (Alfonso-Loeches et al., 2012; Tong, Yu, Deochand, et al., 2015; Tong, Yu, Silbermann, et al., 2015; Yalcin, Nunez, Tong, et al., 2015). However, the longer-term effects of ethanol exposure as demonstrated by neuroimaging are the loss of myelinated fibers together with axonal degeneration (Pfefferbaum & Sullivan, 2002; Wang et al., 2009).

The selective vulnerability of WM in ARBD is not well understood yet evidence suggests that some aspects are partially reversible with abstinence (Bartsch et al., 2007; Estilaei et al., 2001; Gazdzinski et al., 2010; Monnig et al., 2013; Yalcin et al., 2017), whereas other facets are not. For example, abstinence-mediated reversal or attenuation of white matter atrophy was demonstrated in human longitudinal studies using neuroimaging approaches (Bartsch et al., 2007; Estilaei et al., 2001; Gazdzinski et al., 2010; Monnig et al., 2013; Yalcin et al., 2017), and in a chronic ethanol-fed rat model using histopathologic studies (Yalcin et al., 2017). However, persistence of brain pathologies beyond the period of ethanol-exposure reflects a degree of permanence that likely requires additional, targeted interventions. The considerable gap in knowledge is due to inadequate research as shown by the disproportionately low number of publications between 1960 and 2025 that pertain to alcohol’s cellular, biochemical, and molecular pathologic effects on WM (2,134) or the cellular, biochemical and molecular pathologic effects on oligodendrocytes (729) and myelin (477) compared with neurons (23,560). Similarly, alcohol-related brain research publications on astrocyte (1,872) and microglial (978) pathologies also lag. An improved understanding of ARBD’s pathogenesis could lead to novel therapeutic approaches that reduce cognitive impairment and disability.

Studies of white matter pathology in ARBD have been conducted in humans with alcohol use disorders (AUDs) and in experimental models. The human studies included well-characterized subjects evaluated with neurocognitive tests and neuroimaging (Chanraud et al., 2007; Elofson et al., 2013; Fortier et al., 2014; Monnig et al., 2013; Pfefferbaum & Sullivan, 2002; Pop-Jordanova & Demerdzieva, 2022; Schmidt et al., 2005; Wang et al., 2009), or by postmortem brain examination (Chanraud et al., 2007; de la Monte, 1988; Harper et al., 1990). Although the quality of alcohol dose monitoring can vary and is mainly captured through standardized questionnaires, human studies are crucial for understanding AUD effects that may be preventable or therapeutically remediable in people. Postmortem human brain studies are invaluable for corroborating structural abnormalities seen by neuroimaging, but also the biochemical and molecular basis of ARBD for eventual integration with non-invasive biomarker data. On the other hand, experimental models are indispensable for studying alcohol dose and duration effects and incremental effects on neurobehavioral function, neuropathology, and molecular and biochemical integrity. The National Institute on Alcohol Abuse and Alcoholism (NIAAA) endorses several in vivo models to produce pathology or neurobehavioral deficits such as chronic feeding with standardized commercial liquid diets, including the Lieber-DeCarli formula, binge-administration of ethanol by oral gavage or intraperitoneal injection, and vaporized ethanol exposures (Tabakoff & Hoffman, 2000). In addition, in vitro models ranging from primary isolated brain cell cultures or co-cultures to 3-D organoids have been utilized to study the effects of alcohol exposures. A major advantage of in vitro models is that they enable tight control over the experimental conditions. Complementary in vitro, in vivo, and human studies provide the greatest opportunity to understand alcohol-related disease mechanisms and develop novel targets and optimize therapeutic approaches.

Acute Versus Chronic ARBD

Short-term (up to 48 hours) in vivo or in vitro models are used to study acute effects of ethanol exposure on the brain or cultured brain (neuroglial) cells, whereas long-term (days to weeks for experiments, years for humans) in vivo exposures are used to characterize responses reflective of chronic ARBD. The models can be generated by binge (large amounts in short periods), chronic (steady, consistent daily amounts), or chronic plus binge administrations (Bertola, Mathews, et al., 2013; Bertola, Park, et al., 2013). Binge administrations range from 1 to 5 g/kg of ethanol, and chronic exposures are produced by feeding liquid diets containing from 1% (volume/volume (V/V)) to 9% V/V (Piano et al., 2001; Soscia et al., 2006). In addition, the sustained effects of either short- or long-term ethanol exposures can be evaluated beyond the period of ethanol exposure to delineate the permanency of neurotoxic/neurodegenerative effects, including neurodevelopmental abnormalities subsequent to gestational alcohol exposures (de la Monte & Wands, 2010).

Acute alcohol-related brain pathology linked to varying degrees of intoxication is characterized by brain cellular and tissue swelling, particularly in WM, prominence of metabolic (Type 2) astrocytes distributed in both gray and WM structures, and myelin pallor (de la Monte & Kril, 2014). In humans and experimental models, those pathologies are virtually inseparable from concomitant acute alcohol-related liver injury, raising uncertainty about primary versus secondary (hepatic-mediated) encephalopathy. Fortunately, in vitro experiments have clarified that both ethanol and its principal metabolite, acetaldehyde, have direct neurotoxic effects (Lamarche et al., 2004; Tong et al., 2011). Ethanol readily crosses the BBB and is principally metabolized by catalase pathways and cytochrome p450 (Peana et al., 2017; Zakhari, 2006). Therefore, ethanol-mediated damage to the brain is worsened at higher blood levels and durations of exposure, but limited by metabolism to carbon dioxide and water (Zakhari, 2006). Acetaldehyde neurotoxicity in the CNS can be mediated by ethanol metabolism via cytochrome p450 isozymes such as CYP2E1 and catalase-H₂O₂ peroxidase (Peana et al., 2017; Zakhari, 2006), or due to its transport across the BBB from systemic sources (Heap et al., 1995; Peana et al., 2017). Mechanistically, the early effects of ethanol on cultured neuronal cells and brain tissue following in vivo exposures include oxidative injury, lipid peroxidation, activation of pro-apoptosis mechanisms, neuroinflammation, and impairments in cell membrane integrity (de la Monte & Wands, 2001; de la Monte et al., 2000). WM pathology includes injury with altered myelin-associated protein and lipid expression (Gameiro-Ros et al., 2023; Yalcin, Tong, & de la Monte, 2018).

In contrast, the long-term sustained or progressive effects of chronic ethanol exposure/feeding are complex due to multi-pronged mechanisms of tissue injury causing degeneration. Experimental models have been instrumental in characterizing chronic ARBD because they lack superimposed social, cultural, dietary, and environmental insults that could impact disease severity. In the Long Evans rat models of chronic ethanol feeding, WM atrophy, demyelination, myelin degeneration, and axonal damage in the prefrontal region, temporal lobe, and anterior corpus callosum were found to progress over time from 3 to 8 weeks on the 36% caloric (high dose) ethanol-containing Lieber-DeCarli diet (Papp-Peka et al., 2017; Tong, Yu, Silbermann, et al., 2015; Yalcin et al., 2017) (Figure 1). In addition, the neuropathological effects of chronic alcohol exposure were determined to be dose-dependent (Soscia et al., 2006), like observations in humans (Aronson & Hagberg, 1998).

Figure 1.

Experimental alcohol-mediated white matter degeneration. (a-f) Long Evans adult rats were fed Lieber-DeCarli isocaloric liquid diets that contained 0% ethanol (Control) or 36% ethanol (n = 12/group) for up to 8 weeks. Formalin-fixed paraffin-embedded 5-microns (µm)-thick histological sections of anterior corpus callosum (aCC) from (A) control and (B) ethanol-fed rats were stained with Luxol Fast Blue, hematoxylin, and eosin. Myelin is stained blue. Note the striking ethanol-associated atrophy and pallor of myelin staining, especially at 8 weeks. (C, D) Glutaraldehyde-fixed, epon embedded 1 µm-thick section stained with toluidine blue show abundant, densely packed large (arrows) and small (example circled) myelinated axons in (C) control versus reduced densities and degrees of large (arrows) and small (circled) fiber myelination (thin, light staining) in (D) ethanol-exposed white matter. (E,F) High-resolution transmission electron micrographs showing (E) abundant myelinated large and small axons in control and (F) marked reduction of myelinated axons in the ethanol sample. Magnifications: (A) and (B) 200×; (C) and (D) 650×; (E) and (F) 15,000×, final (1 µm scale bars at the lower right of each panel) (Papp-Peka et al., 2017; Yalcin et al., 2017).

In contrast to the acute/short-term abnormalities, chronic alcohol exposures impair performance on neurobehavioral tests together with neuropathologic processes characterized by alterations in myelin lipid composition (Yalcin, Nunez, Tong, & de la Monte, 2015; Yalcin, Nunez, Tong, Cornett, et al., 2015), inhibition of insulin/insulin-like growth factor type 1 (IGF-1) signaling through cell survival and metabolic pathways (Nguyen, Le, Tong, Silbermann, et al., 2012; Tong, Yu, Silbermann, et al., 2015; Xu et al., 2003), increased pro-inflammatory cytokine activation (Crews & Nixon, 2009), oxidative stress, and lipid peroxidation (de la Monte, 2013; de la Monte, Longato, Tong, DeNucci, et al., 2009; Pascual et al., 2003; Tiwari & Chopra, 2013; Yalcin, Tong, Gallucci, et al., 2018) (Figure 2). Oxidative stress, lipid peroxidation, neuroinflammation, and proneness to apoptotic and cytotoxic cell death represent overlapping features of acute and chronic ARBD. The transition from early- to late-stage WM ARBD is marked by atrophy with loss of axons, mature oligodendrocytes, and compact myelin. There is a high likelihood that early-stage ARBD is reversible, including with abstinence (Bartsch et al., 2007; Estilaei et al., 2001; Yalcin, Tong, & de la Monte, 2018) whereas the chronic pathologies are more enduring and progressive, particularly with continued alcohol misuse. Therefore, understanding the critical factors that drive the development of chronic ARBD is a high-priority goal for reducing the socioeconomic and personal burdens linked to the high rates of morbidity and mortality in the USA, Australia, and worldwide. This review discusses how alcohol mediates its injurious and degenerative effects on oligodendrocytes and WM structures, focusing on known mechanisms and relates these responses to the effectiveness of potential therapeutic strategies.

Figure 2.

Chronic experimental alcohol exposure increased indices of oxidative stress and lipid peroxidation. Protein carbonyl and the isoprostane 8-iso PGF2α, produced by the non-enzymatic peroxidation of arachidonic acid in membrane phospholipids, were measured in cerebral tissue from Long Evans rats maintained for 8 weeks on isocaloric liquid diets with 0% (control) or 36% (caloric) ethanol). Protein carbonyl and 8-iso-PGF2α were measured using commercial assay kit reagents (Tong, Yu, Deochand, et al., 2015). Graphs depict the means ± S.D. of results from 8 rats per group. Inter-group differences were evaluated with Student t-tests. Significant P-values are shown within the panels.

ARBD Targeting of Oligodendrocytes

Alcohol’s targeting of WM in mediating acute, subacute, or chronic toxic, metabolic, and degenerative pathologies (de la Monte, 1988; de la Monte & Kril, 2014; Harper et al., 1985; Kril et al., 1989; Kril & Halliday, 1999) is due to its direct injurious effects on, and functional impairments of oligodendrocytes (Benjamins et al., 2011; Creeley et al., 2013). Experimental ex vivo slice culture and in vivo rat models demonstrated that the early effects of alcohol on WM and oligodendrocytes are mediated by cytotoxic injury, oxidative stress, and neuroinflammation, resulting in myelin breakdown and lipid peroxidation (Tong, Yu, Deochand, et al., 2015; Yalcin, Tong, & de la Monte, 2018) (Figure 2). The sustained or progressive WM degeneration resulting from longer-term/chronic exposures is mediated by intrinsic dysfunction of signal transduction, energy metabolism, and gene expression related to myelin proteins and lipids in oligodendrocytes (Tong, Yu, Silbermann, et al., 2015; Tong, Leão, et al., 2017; Yalcin, Nunez, Tong, Cornett, et al., 2015). Myelin is a fatty sheath within the extended membranes of oligodendrocytes wrapped tightly around adjacent axons. Therefore, oligodendrocyte degeneration and death result in myelin loss and reduced brain connectivity (Campagnoni & Macklin, 1988; Collarini et al., 1991; Gordon et al., 1990; Nave & Milner, 1989).

Oligodendrocyte Proteins Modulated by Alcohol Exposure

The membranous protein component of myelin is also synthesized by oligodendrocytes and changes with maturation and function. Oligodendrocytes develop from oligodendrocyte precursor cells (OPC) that differentiate into immature followed by mature myelin-producing oligodendroglia. Mature oligodendroglia express integral membrane proteins including myelin basic protein (MBP), myelin-associated glycoprotein (MAG), myelin oligodendrocyte glycoprotein (MOG), proteolipid protein (PLP) (Bordner et al., 2011) and O4 sulfatide (Back, 2017), as well as adenoma polyposis coli (APC) (Back, 2017). PLP (30 kDa) is the most abundant protein in CNS myelin (Groseclose et al., 2007; Nicklay et al., 2013). The activation of mechanistic target of rapamycin (mTOR) via phosphatidylinositol-3-kinase (PI3K)-protein kinase B (Akt) is essential for oligodendrocyte differentiation and enhancement of stage-specific antigen and myelin protein expression including MBP and PLP (Tyler et al., 2009). mTOR is a component of two multiprotein complexes, mTOR complex 1 and 2 (mTORC1 and mTORC2). The loss of mTORC2 signaling in oligo-progenitor cells reduces APC (mature) and increases PDGFRA (immature) expression in oligodendrocytes. In addition, the inhibition of mTORC2 reduces myelination (Carson et al., 2015), myelin thickness, axonal density, and oligodendrocyte populations, and increases gliosis (Grier et al., 2017). Chronic ethanol-mediated delays in oligodendrocyte maturation are marked by reduced expression of MBP and MAG, and de novo myelin synthesis (Chiappelli et al., 1991; Gnaedinger & Druse, 1984; Gnaedinger et al., 1984; Tong, Andreani, et al., 2016), together with increased expression of immature oligodendrocyte proteins (Tong, Yu, Deochand, et al., 2015; Tong, Andreani, et al., 2016) (Figure 4), possibly due to inhibition of IGF-1 signaling through Akt (Carson et al., 2015).

Figure 4.

Frontal white matter from control and ethanol-exposed rats was used to measure oligodendrocyte mRNA expression using the RT² PCR array system with results normalized to housekeeping genes. Data were analyzed using the SA Biosciences software which automatically calculates ^ΔΔC_t based on fold changes relative to 3 housekeeping genes. Box plots depict the means (horizontal bars), 95% confidence interval limits (upper and lower boundaries of the boxes), and range (upper and lower stems) corresponding to cDNA measured in white matter tissue (n = 4/group). Inter-group comparisons were made using the Student’s t-test (Graphpad Prism 10, Boston, MA). Significant differences are displayed in each panel. Abbreviations: PDGFRA-platelet-derived growth factor receptor type alpha, CNP-cyclic nucleotide phosphodiesterase; PLP1-proteolipid protein 1; RPAIN = replication protein A interacting protein; MOG = myelin oligodendrocyte glycoprotein.

Oligodendrocyte-Myelin Lipids Modulated by Alcohol Exposure

Myelin has a lipid (cholesterol, sphingolipids, and phospholipids) content of 70% to 85% of its dry mass (Schmitt et al., 2015), in contrast to most plasma membranes which contain nearly equal proportions of protein and lipid. Sphingolipids, i.e., sulfatides, cerebrosides, and sphingomyelins, are distributed in the extracellular membrane leaflet and mediate myelin formation, myelin maintenance, and neuronal plasticity (Honke, 2013; Schmitt et al., 2015; Takahashi & Suzuki, 2012). Sphingolipids together with cholesterol form lipid raft microdomains that serve to regulate membrane fluidity, protein trafficking, and signal transduction (Korade & Kenworthy, 2008). Myelin phospholipids including phosphatidylserine, phosphatidylinositol, phosphatidylethanolamine, and phosphatidylcholine are localized along the inner cytosolic leaflet and regulate intracellular signaling and membrane trafficking (Di Paolo & De Camilli, 2006; Fernandis & Wenk, 2007; van Meer et al., 2008). For the most part, studies designed to evaluate the effects of alcohol on WM myelin have utilized either chronic exposure models or human postmortem brain tissue. Consequently, little is known about the short-term, acute effects of ethanol toxicity on myelin lipid expression.

Mass spectrometry and lipidomic approaches have greatly facilitated the characterization of chronic moderate or heavy alcohol exposure effects on WM myelin lipids measured in tissue homogenates/extracts or histological sections. The main achievements of related research that delineated alcohol-related shifts in WM lipid profiles were achieved through advances in accessible implementation of matrix-assisted laser desorption ionization-imaging mass spectrometry (MALDI-IMS) for efficient in situ characterization of myelin lipids (Yalcin & de la Monte, 2015), which eliminates the requirement of generating tissue extracts (Roux et al., 2015; Roux et al., 2016; Wang et al., 2008). A further advance was made via the use of tissue microarrays (TMAs) combined with MALDI-IMS to simultaneously examine multiple samples under the same experimental conditions (Gameiro-Ros et al., 2023).

Using the MALDI-IMS approach, ARBD was shown to be associated with broad qualitative and quantitative alterations in WM phospholipid and sphingolipid expression in both humans and experimental animals (de la Monte et al., 2018; Yalcin, Nunez, Cornett, et al., 2015). Importantly, those studies demonstrated reduced sulfatide and sphingomyelin, and increased ceramide with ARBD (de la Monte, Longato, Tong, DeNucci, et al., 2009; de la Monte et al., 2016; Krotow et al., 2016; Roux et al., 2015; Yalcin, Nunez, Tong, Cornett, et al., 2015) (Figure 3), the relevance of which pertains to the lipotoxic, neuroinflammatory, oxidative stress, and insulin resistance effects of excess ceramide, and neurodegeneration with cognitive impairment linked to reduced sulfatide (Gottfries et al., 1996; Han et al., 2002; Marbois et al., 2000). In addition, chronic alcohol exposures lead to increased generation of ceramide via activation of sphingomyelinases and stress kinases and increased activation of ceramide signaling pathways triggers apoptosis (Pascual et al., 2003).

Figure 3.

Heavy alcohol exposures similarly alter cerebral white matter and O4+ oligodendrocyte sphingolipid profiles, resulting in predominantly reduced sulfatide expression relative to control. Specific sphingolipid species (ceramides-Cer, sphingomyelin-SM, sulfatide-ST, C13 isotopes of lipids, and Lactosylceramide-LacCer) along with their mass/charge ratios (m/z) were detected by MALDI mass spectrometry. The aligned data bar plots corresponding to effects of chronic ethanol exposure (rat model) on the expression levels of the sphingolipids in paired samples of frontal white matter tissue and isolated O4+ oligodendrocytes from the same samples (N = 3/group). Data show calculated mean percentage differences which ranged from −65 (blue) to +58 (red). Note conspicuous parallel results suggesting that oligodendrocyte pathology accounts for many of the alterations in white matter sphingolipid expression following ethanol exposure. C13 refers to an isotope of specific lipids.

Other Cell Types Affected in WM ARBD

Besides oligodendrocytes, responses in neurons, astrocytes, microglia, and possibly endothelial cells contribute to the pathology of ARBD. Alcohol’s adverse effects on the survival and function of CNS neurons have been studied extensively in relation to dose, duration, timing, aging, and development. Neurotoxicity leads to loss of neurons, synaptic connections, and plasticity (de la Monte & Kril, 2014). Neuronal injury leads to WM axonal degeneration (Papp-Peka et al., 2017). The resulting loss of connectivity leads to cognitive-motor deficits (de la Monte & Kril, 2014). Astrocytes, like oligodendrocytes play critical roles in neurotransmission and signal transduction, responding to both alcohol exposure and withdrawal. The latter alters glutamate and GABA metabolism and thereby contributes to neuronal dysfunction, including the transmission of WM axonal signals (Miguel-Hidalgo, 2018). Alcohol-exposed astrocytes compromise neuronal survival and function, particularly during development (Guizzetti et al., 2014). In addition, astrocytes can be targeted by ethanol leading to inhibition of proliferation (Kane et al., 1996), increased generation of glial fibrillary acidic protein (GFAP) (Dalçik et al., 2009), and cell death via ceramide signaling pathways that trigger apoptosis (Pascual et al., 2003).

Ethanol also alters membrane lipid fluidity of astrocytes and oligodendrocytes, suggesting that damage to both cell types contributes to the deficits in myelination (Qu et al., 1999). Chronic ethanol exposure promotes microglial activation via induction of pro-inflammatory cytokines (Crews & Nixon, 2009; Drew et al., 2015). Such microglial activation increases oxidative stress, driving neuronal injury and dysfunction (Guizzetti et al., 2014), and stimulates the chemoattractant MCP-1/CCL2 which promotes neurotoxicity leading to brain atrophy (Zhang & Luo, 2019). Finally, a relatively scant body of literature has provided evidence that ethanol exposures have pathogenic effects on vascular cells including endothelium during development and in the mature brain (Bukiya et al., 2016; Kobayashi et al., 1996; Momin et al., 2023). Ethanol’s toxic and degenerative effects on the cerebral vasculature have largely emphasized loss of BBB integrity resulting in increased permeability and leakage (Persson & Rosengren, 1977; Rosengren et al., 1977) which could account for brain swelling in the acute stages of intoxication (de la Monte & Kril, 2014). Ethanol exposures during development can have substantial adverse effects on angiogenesis rendering brain structures deficient in microvascular networks (Jégou et al., 2012) and therefore prone to hypoperfusion. Altogether, the findings in these studies highlight ethanol’s effects on WM astrocytes, microglia, vasculature, and neurons, and thereby illustrate the complexity of their interrelated toxic and degenerative effects on oligodendrocytes.

Insulin/IGF-1/mTOR Support of Oligodendrocyte Functions

Growing evidence supports the concept that insulin and IGF-1 signaling through downstream pathways that involve PI3K, Akt, and the mechanistic (mammalian) target of rapamycin (mTOR) regulates and maintains oligodendrocyte functions, including cell survival, metabolism,′ homeostasis, myelination, and myelin integrity (Figlia et al., 2018; Narayanan et al., 2009). Alcohol has crippling effects on the insulin/IGF-1-Akt-mTOR pathways in the brain, resulting in significant inhibition of target gene expression and function across the lifespan (Andreani et al., 2016; de la Monte & Wands, 2002; Ewenczyk et al., 2012; Tong, Yu, Deochand, et al., 2015; Tong et al., 2022). In WM, ethanol’s adverse effects are linked to impairments in PI3K-Akt-mTOR signaling (de la Monte, Elgas, et al., 2023; Papp-Peka et al., 2017; Tong, Yu, Deochand, et al., 2015; Tong et al., 2022). This is reflected in a reduced ability of oligodendrocytes to synthesize and maintain compact myelin, which is essential for optimum neurotransmission and protection of axons and dendrites. It will be important to extend this knowledge by characterizing ethanol’s effects on insulin and IGF-1 signaling through Akt-mTOR-mTOR and mTORC1 and mTORC2 in WM due to the critical roles these pathways play in maintaining oligodendrocyte survival and function (Broughton et al., 2007; D’Ercole & Ye, 2008; de la Monte & Wands, 2005; Freude et al., 2008; Folli et al., 1996; Gammeltoft et al., 1985; Hill et al., 1986; Unger, Livingston, et al., 1991; Unger, Moss, et al., 1991). The mTORC1 and mTORC2 complexes respectively include the regulatory-associated protein of mTOR (raptor) or the rapamycin-insensitive protein (rictor), for modulating a broad range of functions such as nutrient sensing, protein synthesis, cell proliferation, cell migration, cytoskeletal remodeling. Emphasis on Akt-mTOR → mTORC is justified due to its established roles in myelin synthesis and maintenance (Dubois et al., 2014; Figlia et al., 2018; Laplante & Sabatini, 2012b; Narayanan et al., 2009), and recent data from models of brain insulin resistance (de la Monte, 2017; Lee et al., 2017).

Insulin and IGF-1 signaling support oligodendrocyte survival and myelin-associated functions (Barres et al., 1993; Chesik et al., 2008; Freude et al., 2008; Gong et al., 2008), largely through the PI3K-Akt pathway. Detailed analysis of this pathway using mRNA and protein-based studies revealed early or short-term ethanol inhibitory effects on signaling protein phosphorylation, but with chronic exposures, combined effects of reduced protein phosphorylation and inhibition of signaling protein expression (mRNA and protein) accounted for the impairments in survival and metabolic functions in WM and oligodendrocytes (Cohen et al., 2007; de la Monte et al., 2008; Nguyen, Le, Tong, Mellion, et al., 2012; Nguyen, Le, Tong, Silbermann, et al., 2012; Tong, Yu, Silbermann, et al., 2015). Although previous studies demonstrated roles for both reduced kinase and increased phosphatase activities as acute or short-term mediators of ethanol-impaired insulin/IGF-1 signaling, further studies are needed to assess their relative contributions to signaling impairments identified following prolonged ethanol exposure.

Deeper Dive into mTOR Signaling Impairments in ARBD

The mTOR pathways mediate their effects on cellular functions by activating mTOR protein complex 1 (mTORC1) or mTORC2 (Tyler et al., 2009; Zhou & Huang, 2010). mTORC1, which includes raptor (Tyler et al., 2009) and is sensitive to Rapamycin inhibition, stimulates oligodendrocytes to produce myelin, whereas mTORC2, which includes rictor (Tyler et al., 2009) and is largely resistant to rapamycin, signals through p70S6K to modulate energy metabolism (Dubois et al., 2014). Correspondingly, rapamycin reduces WM volume by inhibiting oligodendrocyte survival and maturation, myelin synthesis, and myelin maintenance, whereas inhibition of mTOR → mTORC2 reduces S473-Akt and phospho-p70S6K, critically altering cell survival and metabolism (Figlia et al., 2018; Goebbels et al., 2010; Narayanan et al., 2009).

Inhibition of signaling through mTORC1 could account for ethanol-driven deficits in myelinogenesis and myelin maintenance (Dubois et al., 2014; Figlia et al., 2017; Laplante & Sabatini, 2012a; Narayanan et al., 2009), whereas its inhibition through mTORC2 could account for impairments in oligodendrocyte maturation (Dubois et al., 2014). Multiplex magnetic bead-based immunoassays of total and phosphorylated insulin receptor (R), IGF-1R, IRS-1, Akt, GSK-3β, p70S6K, PRAS40, and PDK (Longato et al., 2012; Ramirez et al., 2014; Tong et al., 2010; Tong et al., 2009), together with bead-based immunoassays of total and phosphorylated mTOR-mTORC1/mTORC2 were used to demonstrate prominent ethanol inhibition of signaling through mTOR, adversely impacting both mTORC1 and mTORC2 in WM (de la Monte, Tong, et al., 2023) (Figure 5). Therefore, disruption of either mTOR pathway would likely lead to myelin breakdown, stress, lipid peroxidation, ceramide generation, and further impairment of insulin/IGF signaling.

Figure 5.

Ethanol inhibition of white matter insulin receptor and Akt phosphorylation and mTOR immunoreactivity in frontal lobe tissue (n-4/group). Immunoreactivity to total insulin receptor (Insulin R) protein, active, phosphorylated (pY1162/1163) insulin R, Akt, 473S-phosphorylated Akt, and mTOR were measured by multiplex magnetic bead-based ELISA. The calculated relative levels of phosphorylated (p/T) insulin R and Akt were compared between control (Con) and ethanol (EtOH) exposed samples. Results are depicted with box plots (see Figure 4’s legend for details). Inter-group statistical comparisons were made by student’s t-test analysis using Graphpad Prism 10 (Boston, MA). Significant p-values are shown within the panels.

Paradoxically, other studies have demonstrated adverse effects of mTOR activation such that microglial proliferation and the generation of reactive oxygen species and pro-inflammatory cytokines in states of acute oxidative stress, hypoxia, ischemia, or glucose deprivation (Dello Russo et al., 2009; Hu et al., 2020; Karunakaran et al., 2019), and following repeated binge alcohol exposures (x3 over 7 days) (de la Monte, Tong, et al., 2023). Furthermore, blockade of mTORC1 may attenuate ethanol-associated neurobehavioral deficits (Li et al., 2016) and the inhibition of mTOR can be neuroprotective against inflammatory (Srivastava et al., 2016) and ischemia/reperfusion injury (Wang et al., 2020). Given the complexity of cell types and cellular responses to insulin/IGF signaling through PI3K-Akt-mTOR and mTORC1 versus mTORC2, together with the impacts of different durations and levels of ethanol exposure, additional studies are needed to delineate the circumstances and factors that drive positive versus negative outcomes. The latter would be most relevant to our understanding of mechanisms that drive transitions from short-term reversible to long-term-sustained or progressive pathologies in WM ARBD.

Roles of Impaired Insulin/IGF Signaling in WM ARBD

One aspect of ARBD clearly associated with chronic ethanol consumption is impaired signaling through both the insulin and IGF-1 receptors as mediators of neuronal degeneration and cognitive-behavioral deficits (Cohen et al., 2007; Corl et al., 2005; de la Monte, Longato, Tong, DeNucci, et al., 2009; de la Monte, 2013; de la Monte et al., 2008; Lindtner et al., 2013; Tong, Yu, Silbermann, et al., 2015). Evidence now shows that these same pathways have important roles in WM pathology (Andreani et al., 2014; de la Monte, Longato, Tong, DeNucci, et al., 2009; Tong, Yu, Silbermann, et al., 2015; Tong, Gonzalez-Navarrete, et al., 2017), and particularly through IGF-1’s effects on oligodendrocyte function. For example, earlier studies showed that in transgenic mice, overexpression of IGF-1 increased brain size, oligodendrocyte abundance, and myelin content (Carson et al., 1993; D’Ercole et al., 1996), whereas depletion of IGF-I or IGF-I receptor genes, or over-expression of IGF binding proteins impaired brain growth (D’Ercole et al., 2002). More information about how impairments in insulin versus IGF-1 signaling impact WM pathology in ARBD is needed since the growth and metabolic effects of these trophic factors may differ by brain region, developmental stage, aging, and dose/duration of alcohol exposure. Furthermore, the differential effects of alcohol on insulin versus IGF-1 signaling may be relevant to the transition from acute/subacute to chronic/progressive ARBD.

Beyond its direct toxic injury and degenerative effects on neurons and oligodendrocytes, alcohol exposures mediate ARBD by causing oxidative stress and promoting the generation of pro-inflammatory cytokines (de la Monte & Kril, 2014; Dennis et al., 2014). In a small study of human postmortem brains, we detected significantly elevated levels of several pro-inflammatory cytokines in anterior frontal lobe tissue from chronic alcoholics compared with non-alcoholic controls (Table 1). The predominantly elevated pro-inflammatory cytokine responses could lead to oligodendrocyte injury and functional impairments and thereby account for the altered WM lipid profiles and myelin integrity observed in ARBD (de la Monte & Kril, 2014; de la Monte et al., 2016; de la Monte et al., 2018). Furthermore, since the pro-inflammatory cytokine TNF-α has both neuroprotective and neurodegenerative actions in the brain (Abd-El-Basset et al., 2021; Suzuki et al., 2004), its seemingly paradoxical reduction in AD brains could reflect a decline in neuroprotection, corresponding with the likely effects of the more broadly increased pro-inflammatory cytokines (de la Monte & Kril, 2014; de la Monte et al., 2018; de la Monte et al., 2016).

Table 1.

Increased brain pro-inflammatory cytokine levels in human chronic alcoholics.

Human frontal WM	IL-1β	TNF-α	IL-6	IFN-γ	IL-16
Control	12.9 ± 0.3	10.2 ± 0.3	10.0 ± 0.5	5.0 ± 0.2	289 ± 51
Alcoholic	20.9 ± 2.3	8.7 ± 0.5	15.9 ± 1.4	6.8 ± 0.4	285.3 ± 40.5
T-test	P = 0.001	P = 0.01	P = 0.003	P = 0.0009	NS

Human adult fresh postmortem frontal lobe white matter (WM) tissue samples from 8 controls and 8 with alcohol use disorder were analyzed for pro-inflammatory cytokine expression. Immunoreactivity was measured with a commercial magnetic bead-based ELISA platform. The control group had a mean age (±S.D) of 54.4 ± 6.9 years and included 5 males and 3 females. The alcoholic group had a mean age of 54.8 ± 7.2 years and included 5 males and 3 females. Values reflect arbitrary fluorescence light units normalized to protein concentration (mean ± S.D.). Inter-group statistical comparisons were made by student’s t-test analysis. Abbreviations: IL = interleukin; TNF-α = tumor necrosis factor-alpha; IFN-γ = interferon-gamma; NS = not significant.

Inflammation and oxidative stress with attendant accumulation of neurotoxic ceramide compromise insulin/IGF-1 signaling through Akt (de la Monte & Kril, 2014) and reduce mTOR activation (Dubois et al., 2014; Figlia et al., 2017; Laplante & Sabatini, 2012; Narayanan et al., 2009) which is needed for oligodendrocyte and neuronal functioning. Recently, we validated this concept with an ex vivo frontal lobe slice culture model that we used to measure total (T) and phosphorylated (P) signaling molecules in the upstream, mid-level, and downstream components of the pathway through mTORC1 and mTORC2 (de la Monte, Elgas, et al., 2023). The study was conducted with commercial 11-Plex magnetic bead-based enzyme-linked immunosorbent assay (ELISA) reagents (Millipore Sigma, Bedford, MA). Results corresponding to the calculated relative levels of phosphorylated insulin receptor (Insulin R) (^pY1162/1163-Insulin R/Total Insulin R) and ^473S- Akt/Total, and total mTOR are shown in Figure 5. Those studies showed that ethanol exposures impaired insulin/IGF-1/Akt-mTOR-mTORC1/2 signaling in WM, and compromised oligodendrocyte function by increasing the expression of immature (PDGFRA, Vimentin, CNP) and decreasing the expression of mature (PLP, RPAIN, MOG) myelin oligodendrocyte proteins (Figure 4) and reducing the relative abundance of sulfatide compared with ceramide (Figure 3) (Homans et al., 2022; Yalcin, Tong, & de la Monte, 2018). These responses are likely important mediators of WM atrophy and degeneration in ARBD (Figure 5).

Complexity of Pathogenic Factors in ARBD

Toxic effects of ethanol leading to cellular injury with membrane damage, altered membrane lipid composition, inflammation, and oxidative stress with lipid peroxidation have been well-characterized in liver (Deaciuc et al., 2000; Baraona et al., 2002; Carmiel-Haggai et al., 2003; Lischner et al., 1971; Teplova et al., 2010). Furthermore, in liver, progression from the acute stages of alcohol-induced cytotoxic injury to chronic progressive ARLD and degeneration that culminate in cirrhosis and liver failure (Lischner et al., 1971; Davidson, 1981; Altamirano & Bataller, 2011; Diehl, 1997) has been linked to insulin resistance (de la Monte, Longato, Tong, DeNucci, et al., 2009; de la Monte et al., 2012; Longato et al., 2012; Patel et al., 1991; Ramirez et al., 2013; Ronis et al., 2007). While acute alcohol-related liver injury can be markedly reduced or cured by abstinence (Estilaei et al., 2001; Brunt et al., 1974; Testino & Borro, 2012), chronic ARLD requires intervention and is responsive to insulin sensitizers (Pang et al., 2009) or the ceramide inhibitor, myriocin (Tong et al., 2014). In contrast, less is known about the distinctions between acute and chronic alcohol-related CNS pathology due to relatively limited research (Melgaard, 1983; Ke et al., 2011). However, like liver, the acute alcohol-medicated CNS effects and pathology are associated with increased inflammation, oxidative stress, and lipid peroxidation with a build-up of malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) (Tong et al., 2011). In addition, ethanol disrupts CNS cell membrane integrity (Crews & Nixon, 2009; de la Monte & Wands, 2002; de la Monte, Longato, Tong, DeNucci, et al., 2009; Le et al., 2013; Lindtner et al., 2013; Marcondes et al., 2008; Nguyen, Le, Tong, Silbermann, et al., 2012; Tiwari & Chopra, 2013; Tiwari et al., 2010; Xu et al., 2003) and dysregulates lipid metabolism which is marked by reduced sulfatide and increased toxic ceramide (Bae et al., 2014; de la Monte, 2013; de la Monte, Longato, Tong, DeNucci, et al., 2009; Pascual et al., 2003; Roux et al., 2015). Eventually, as in liver, ethanol impairs insulin and IGF signaling through Akt metabolic and cell survival pathways in the CNS (de la Monte et al., 2008; de la Monte et al., 2012; de la Monte, Longato, Tong, & Wands, 2009; Ronis et al., 2007). How these pathogenic factors drive the transition from acute to subacute or chronic stages of WM ARBD in which myelin and oligodendrocyte degeneration worsen (Ayala et al., 2014) and DNA damage (8-OHdG) (de la Monte & Wands, 2005; Soscia et al., 2006) increases, is poorly understood and deserves further study. However, the findings in previous and preliminary studies suggest that reducing oxidative stress is important. For example, Figure 6 demonstrates significantly higher frontal lobe levels of malondialdehyde adduct (MDA) over an 8-week period of ethanol feeding relative to samples from control diet-fed rats (n = 8/group). MDA was measured in tissue homogenates by ELISA. However, a 2-week period of abstinence after 6 weeks of chronic ethanol diet feeding reduced frontal lobe MDA to levels that were comparable to control. In that same model, abstinence also diminished atrophy and nearly normalized sphingolipid expression in WM (Yalcin et al., 2017). In addition, another study showed that treatment with myriocin, a serine palmitoyltransferase (SPT) inhibitor that reduces pro-inflammatory cytokine-mediated oxidative stress (Dasgupta & Ray, 2017) and oligodendrocyte degeneration (Miller et al., 2017), prevents alcohol-mediated WM atrophy and neurobehavioral dysfunction (Homans et al., 2022). These observations suggest that CNS indices of oxidative stress following chronic ethanol exposures can be favorably modulated by therapeutic interventions including abstinence (as shown in Figure 6) or the use of antioxidants to reduce harm in ARBD.

Figure 6.

Chronic ethanol feeding increased frontal lobe white matter malondialdehyde (MDA) adduct and reactive oxygen species. Abstinence for 2 weeks after 8 weeks of ethanol feeding normalized white matter levels of MDA (Et-R; green symbol). MDA was measured by ELISA in tissue homogenates. N = 8 rats/group. Results are shown with box plots (see legend to Figure 4). Inter-group comparisons were made by student’s t-test analysis using Graphpad Prism 10 software (Boston, MA). *P < 0.05; **P < 0.01; ***P < 0.001.

Over-Arching hypothesis-Dual Stages of Alcohol-Related Brain Disease (ARBD)

Neurobehavioral and cognitive disorders caused by alcohol misuse are tied to CNS pathologies that occur at subcellular, cellular, tissue, organ, and system levels. Primary acute ARBD is manifested by intoxication or hepatic encephalopathy whereas secondary acute ARBD includes head trauma, hepatic myelopathy, and hepato-cerebral degeneration. Chronic ARBD states are heterogeneous and include cognitive-motor impairments and dementia associated with atrophy of the cerebral cortex, diencephalon, WM, cerebellum, hippocampi, or subcortical nuclei and Wernicke’s encephalopathy. The distinct clustering of clinical and neuropathologic features in acute versus chronic ARBD draws attention to the concept that their underlying mechanisms may differ due to pathogenic mechanisms that shift with duration and levels of ethanol exposure.

We hypothesize that the adverse effects of ethanol on WM occur in two major stages, acute and chronic, like the patterns of injury in ARLD (Scaglioni et al., 2011; Seth et al., 2011). In ARLD, short-term exposures mainly cause hepatic steatosis or steatohepatitis which can be reversed by abstinence whereas chronic injury and degeneration lead to cirrhosis and end-stage disease which are not reversed by abstinence (Davidson, 1981; O’Shea & McCullough, 2005). In Long Evans rats, deficits in spatial learning and memory that develop within 3 weeks of chronic ethanol feeding (36% Lieber DeCarli) (Papp-Peka et al., 2017; Tong, Yu, Deochand, et al., 2015) are associated with WM demyelination and relative preservation of axons (Papp-Peka et al., 2017; Tong, Yu, Silbermann, et al., 2015). These effects are associated with increased oxidative stress and neuroinflammation (Papp-Peka et al., 2017; Tong, Yu, Deochand, et al., 2015) marked by activation of astrocytes and microglia and reduced mature myelin-associated glycoprotein expression (Tong, Yu, Silbermann, et al., 2015). Further preliminary studies of isolated oligodendrocytes and WM slice cultures showed that the earliest (24 hours–48 hours) neurotoxic effects of ethanol include increased ceramide and reduced sulfatide (Figure 3) lipid peroxidation, and oxidative stress. Altogether, the evidence suggests that short-term damage to myelin may occur via toxic, metabolic, and inflammatory injury and that oligodendrocytes and myelin are targeted. The accompanying breakdown of myelin may lead to increased levels of cytotoxic ceramide, that exacerbate ethanol’s toxic, oxidative stress, metabolic, and inflammatory effects. An improved understanding of the mechanisms of early potentially reversible WM pathologies could help strategize approaches for preventing long-term neurocognitive deficits.

We propose that, like ARLD, the second chronic and progressive stage of WM degeneration is mediated by inhibition of insulin and IGF-1 signaling through PI3K-Akt-mTOR→ mTORC1/2 with sustained impairments in oligodendrocyte survival and function, leading to inadequate mature myelin maintenance (Figure 4). Damaged myelin exposes axons to a cytotoxic environment, rendering them vulnerable to degeneration. Impairments in insulin/IGF signaling may cause WM degeneration and cognitive-behavioral dysfunctions to progress due to inhibition of enzymes/genes needed to support sulfatides and prevent ceramide accumulation. These long-term effects of ARBD, like ARLD may be remediated by insulin sensitizer (de la Monte & Wands, 2010; Le et al., 2013) and ceramide inhibitor (Homans et al., 2022; Tong et al., 2014; Yalcin et al., 2022) treatments, as suggested by the findings in previous studies. Peroxisome-proliferator-activated receptor (PPAR) agonists and sphingosine 1-phosphate receptor modulator and functional antagonists such as myriocin are respectively important candidate insulin sensitizer and ceramide inhibitor treatments for ARBD (see below). These concepts imply that ARBD treatments should be geared to disease stage. In addition, a fuller understanding of the key molecular and biochemical pathologies associated with early versus late WM ARBD could help delineate stage-appropriate treatments.

The clinical relevance of our two-stage hypothesis is that intervention strategies should be tailored to mechanisms rather than one size fits all. Furthermore, to efficiently transfer laboratory research to clinical practice, non-invasive, sensitive tools are needed to report early versus late-stage WM ARBD pathology. Preliminary studies indicate alcohol exposure effects on WM myelin protein and lipid expression are detectable in ECVs isolated from oligodendrocytes (as shown in Figure 7). Since ECVs are nanoparticles that can be isolated from peripheral blood, it should be possible to detect alcohol exposure effects and responses to treatment in WM-oligodendrocyte ECVs are isolated from plasma. The hope is that with refinements this strategy could lead to the development of rapid, sensitive, non-invasive assays for detecting and monitoring WM ARBD and responses to treatment.

Figure 7.

Potential use of extracellular vesicles (ECV) derived from brain white matter to detect alcohol-related biochemical pathology. These studies were conducted with frontal lobe slice cultures generated from control (C) and chronic ethanol-fed (3 weeks) (Et) Long Evans rats. ECVs were isolated from culture supernatants (exosome-free medium) using the ExoQuick ULTRA EV isolation system (System Bioscience, Palo Alto, CA). ELISAs demonstrated (immunoreactivity in all preparations with elevated levels of CD63 in ECVs from ethanol exposed brains. Duplex ELISAs with results normalized to HSP70 were used to compare C and Et ECV levels of A) CD9, CD63, and CD81 tetraspanins, B) MAG, C) PLP, D) MBP, E) PDGF, and F) GFAP and the findings in O4+ oligodendrocytes and white matter tissue. N = 4 brain slice cultures/group. Data were graphed using box plots and analyzed with Graphpad Prism 10 software (Boston, MA). Inter-group comparisons were made by student’s t-test analysis. *P < 0.05; ***P < 0.001; ****P < 0.0001. The p-value in Panel B corresponds to the calculated statistical trend.

Therapeutic Intervention-Proof of Concept Studies (Figure 8)

Figure 8.

Hypothesis concerning the two-stage development of ARBD, including prominent mechanisms of damage. Differential strategies for treatment should be based on the mechanisms of tissue injury. Abstinence per se in the late stages of ARBD would be clinically inadvisable as a sole therapeutic measure.

Following short-term alcohol exposures, abstinence holds the greatest promise for harm reduction (Yalcin et al., 2017; Yalcin, Tong, & de la Monte, 2018), provided the blood alcohol levels are neither acutely toxic nor fatal. In an experimental rat model, short-term abstinence more effectively reversed early- compared with late-stage ARBD, including spatial learning and memory deficits and myelin loss (Yalcin et al., 2017). The potential benefits of treating early-stage ARBD with antioxidants, anti-inflammatory agents, prednisolone, or N-acetylcysteine, as used for acute alcoholic hepatitis, often accompanied by encephalopathy (Chacko et al., 2011; De et al., 2009; Nguyen-Khac et al., 2011), have not been evaluated. Instead, interventions have focused primarily on reducing liver damage and secondarily on toxic metabolic encephalopathy.

The linking of long-term WM ARBD effects to mechanisms of cellular and tissue injury provides opportunities to treat based on the need to resolve specific cellular dysfunctions. The promising avenues of intervention pertain to the dysregulated sphingolipid metabolism and impairments in insulin and IGF-1 signaling through PI3K-Akt-mTOR (de la Monte et al., 2011; Lizarazo et al., 2013; Pang et al., 2009; Ramirez et al., 2012; Setshedi et al., 2011; Tong et al., 2014). The potential therapeutic approach of normalizing sphingolipid expression was partially addressed using myriocin in experimental in vivo rat models of chronic alcohol feeding (Homans et al., 2022; Lizarazo et al., 2013; Tong et al., 2014; Yalcin et al., 2022). Myriocin, an inhibitor of serine palmitoyltransferase (SPT) that disrupts ceramide synthesis, is a United States Food and Drug Administration (FDA)-approved drug under the names, “ISP-1” and “FTY720” (Penton-Arias & Haines, 2016). Myriocin has been shown to be effective in reversing experimental steatohepatitis caused by alcohol or obesity with metabolic syndrome (Lizarazo et al., 2013; Tong et al., 2014; Yalcin et al., 2022), and WM inflammatory demyelinating disease (Dasgupta & Ray, 2017; Miller et al., 2017). More recently, myriocin was demonstrated to restore alcohol-related neurobehavioral dysfunction and some aspects of the WM myelin lipid-biochemical pathology in an experimental Long Evans rat model of chronic alcohol feeding (Homans et al., 2022). Although promising, CNS recovery was incomplete, indicating that additional co-interventions will be needed to fully prevent or treat chronic progressive ARBD.

Peroxisome-proliferator-activated receptor (PPAR) agonists function as insulin sensitizers and anti-inflammatory agents, activating nuclear receptors to promote lipid metabolism, glucose utilization, and insulin-responsive gene expression (Draznin, 2006; Jiang & Zhang, 2005), and reduce ceramide (de la Monte et al., 2011; Enomoto et al., 2003; Pang et al., 2009; Tomita et al., 2004). Additional effects of PPAR agonists include enhanced neuroprotection via increased brain derived neurotrophic factor expression, increased mitochondrial metabolism, mitochondrial biogenesis, and neurotransmission (Wójtowicz et al., 2020). In the brain, PPAR-δ receptors are abundantly expressed (Hall et al., 2008; Tyagi et al., 2011), followed by PPAR-γ and then PPAR-α, highlighting the importance of therapeutic target design to achieve the intended functional outcomes. PPAR-δ agonists promote myelination and target brain insulin resistance, including that caused by developmental ethanol exposures (de la Monte et al., 2006). PPAR-δ and PPAR-γ but not PPAR-α agonists improve Morris water maze performance in developmental ethanol and toxin-mediated brain insulin resistance models (de la Monte & Wands, 2010; de la Monte et al., 2006). Recently, T3D-959, a novel oral hybrid PPAR δ/γ agonist with 80% δ/20% γ activity (Tong, Deochand, et al., 2016), was shown to reverse WM atrophy and degeneration (de la Monte, 2017; Tong, Dominguez, et al., 2016) and reduce CNS neuroinflammation and oxidative stress in streptozotocin models of brain insulin resistance (de la Monte, 2017).

PPAR agonists that target the δ and γ receptors can reverse experimental ARLD, and also abrogate the CNS structural and functional pathologies in chronic WM ARBD (de la Monte, Longato, Tong, DeNucci, et al., 2009; Le et al., 2013; Tong, Yu, Deochand, et al., 2015). Correspondingly, the PPAR-δ + γ agonists prevented or significantly reduced the spatial learning and memory deficits, myelin loss, and impairments in CNS insulin/IGF signaling in late-stage ARBD (de la Monte et al., 2011), and increased oligodendrocyte maturation (Simonini et al., 2010). Subsequent extensions of these studies utilized dietary soy to provide a natural source of insulin sensitizers instead of the pharmaceutical PPAR-δ and PPAR-γ agonists (de la Monte, Elgas, et al., 2023; Tong et al., 2022). The outcomes were comparable and suggested economically feasible public health measures could be employed to reduce the long-term burdens of chronic WM ARBD.

Conclusions

• Novel dual disease stage hypothesis for ARBD: Heavy alcohol exposures initially cause reversible oligodendrocyte and WM pathology due to oxidative stress, inflammation, and lipid peroxidation. Persistent exposures drive progressive WM degeneration due to the inhibition of PI3K-Akt-mTOR-mTORC signaling in oligodendrocytes.

• Mechanistic hypotheses: Inhibition of Akt-mTOR-mTORC1/2 causes progressive WM ARBD due to: (a) failed maturation and death of oligodendrocytes, compromising myelin integrity; and (b) altered expression of oligodendrocyte genes/enzymes, resulting in sulfatide depletion and ceramide accumulation with cognitive impairment.

• Forward-looking treatment interventions could potentially be based on the dual disease stages of ARBD: Abstinence, insulin sensitizers, and ceramide inhibitors have differential disease-modifying effects on WM ARBD: (a) Early-stage WM ARBD can be remediated by abstinence, anti-inflammatory/antioxidant agents, and/or ceramide inhibitors. (b) Late-stage WM ARBD responds to insulin sensitizers and ceramide inhibitors, but not abstinence. More research is needed to justify these proposed approaches.

Disclosure statement

No potential conflict of interest was reported by the authors.