Moderate blood alcohol and brain neurovulnerability: Selective depletion of calcium-independent PLA2, omega-3 docosahexaenoic acid and its synaptamide derivative as a potential harbinger of deficits in anti-inflammatory reserve

Abstract

BACKGROUND:

Repetitive, highly elevated blood alcohol (ethanol) concentrations (BACs) of 350–450 mg/dl over several days cause brain neurodegeneration and coincident neuroinflammation in adult rats that is localized in hippocampus (HC), temporal cortex (esp. entorhinal cortex; ECX), and olfactory bulb (OB). The profuse neuroinflammation involves microgliosis, increased pro-inflammatory cytokines, and elevations of Ca⁺²-dependent phospholipase A2 (cPLA2) and secretory PLA2 (sPLA2) that both mobilize pro-inflammatory ω–6 arachidonic acid (ARA). In contrast, Ca⁺²-independent PLA2 (iPLA2) and anti-inflammatory ω–3 docosahexaenoic acid (DHA), a polyunsaturated fatty acid believed to be primarily regulated by iPLA2, are diminished. Furthermore, supplemented DHA exerts neuroprotection. Given uncertainties about possible effects of relatively common lower circulating levels during short term binges, we examined how moderate BACs affected the above inflammatory events, and the impact of supplemented DHA.

METHODS AND RESULTS:

Young adult male rats sustaining upper-moderate BACs (~150 mg/dl) from once-daily alcohol intubations were sacrificed with appropriate controls after one week. The HC, ECX and OB were quantitatively examined using immunoblotting, neurodegeneration staining, and lipidomics assays. Whereas neurodegeneration, increases in cPLA2 IVA, sPLA2 IIA and ARA, and microglial activation were not detected, the HC and ECX regions demonstrated significantly reduced iPLA2 levels. Levels of DHA and synaptamide, its anti-inflammatory N-docosahexaenoylethanolamide derivative, also were lower in HC, and DHA supplementation prevented the iPLA2 decrements in HC. Additionally, adult mice maintaining upper-moderate BACs from limited alcohol binges had reduced midbrain iPLA2 levels.

CONCLUSIONS:

The apparently selective depletion by moderate BACs of the metabolically-linked anti-inflammatory triad of hippocampal iPLA2, DHA and synaptamide, as well as of iPLA2 in the ECX, potentially indicates an unrecognized deficit in brain anti-inflammatory reserve that may be a harbinger of regional neurovulnerability.

INTRODUCTION

Over 15 million Americans suffer or have recently suffered from an alcohol (ethanol) use disorder (AUD) (National Institute on Alcohol Abuse and Alcoholism 2017). Chronic, binge AUD causes brain structural (neuronal and synaptic) damage (Zahr et al. 2011) that can result in cognitive impairment and early, sometimes lasting, dementia (Cheng et al. 2017; Sabia et al. 2018). Adult rodent models subjected to repetitive binge intoxication bouts have been often used for neurobehavioral and neuropathological studies. One of the earliest with severely-binged adult (originally male) rats was initially developed for the study of withdrawal seizures (Majchrowicz 1975) and was determined later to cause neurodegeneration in olfactory memory-associated brain regions—notably, dentate granule (DG) neurons of the hippocampus (HC), entorhinal cortex (ECX) pyramidal cells, and olfactory bulb (OB) neurons (Switzer et al. 1982; Collins et al. 1996; Corso et al. 1998). In that model, blood alcohol concentrations (BACs) from ~3 daily binges often exceeded 400 mg/dl, approximating BACs attained in chronic AUD individuals (Majchrowicz and Mendelson 1970; Roberts & Dollard 2010).

It is now known that persistent neuroinflammation is involved in mediating and/or facilitating neurodegeneration in many neurodegenerative diseases and toxic insults, including chronic AUD (Alfonso-Loeches & Guerri 2011; Crews et al. 2013). Neuroinflammatory changes associated with neurodegeneration caused by high binge BACs include microgliosis and enhanced microglial activation indicators, proinflammatory cytokines and chemokines, poly(ADP-ribose) polymerase, and phospholipase A2 (PLA2) enzymes, particularly those regulating ω–6 arachidonic acid (ARA) (Crews et al. 2013; Tajuddin et al. 2014; Marshall et al. 2016).

As critical membrane regulatory enzymes, PLA2 family members catalyze hydrolysis at the sn-2 position of membrane glycerophospholipids, typically yielding a lysophospholipid and a polyunsaturated fatty acid (PUFA) such as ARA or ω–3 docosahexaenoic acid (DHA). The phospholipases exist as multiple isoenzymes within superfamilies that are in three main groups: cytosolic PLA2 (cPLA2), secreted PLA2 (sPLA2)—both being Ca⁺²-dependent—and Ca⁺²-independent PLA2 (iPLA2) (Mouchlis and Dennis 2018). Among cPLA2 and sPLA2 superfamilies, the isoenzymes cPLA2 IVA and, to a considerable degree, sPLA2 IIA, have high affinities for liberating/mobilizing ω–6 ARA (Mouchlis and Dennis 2018). Nonesterified ARA can be autoxidized to generate reactive oxygen species, and/or further metabolized via predominantly pro-inflammatory, pro-oxidative stress pathways, e.g., cyclooxygenases, that engender prostaglandins and thromboxanes, and/or lipoxygenases, producing leukotrienes (Jensen et al. 2009; Sun et al. 2014; Innes and Calder 2018).

Unlike Ca⁺²-dependent PLA2s, iPLA2 appears to be relatively “DHA-specific”, preferentially cleaving the pro-survival ω–3 fatty acid from brain membrane phospholipids (Green et al. 2008). Nonesterified DHA can be further metabolized by cyclooxygenases and/or lipoxygenases to produce metabolites that often exert anti-inflammatory and/or neuroprotective actions (Layé et al. 2018; Sun et al. 2018). For example, a lipoxygenase-derived, hydroxylated DHA metabolite, neuroprotectin D1 (NPD1), protects against ischemic stroke and traumatic brain injury in animal models (Bazan 2018). Also, cytochrome P450-mediated epoxidation of DHA (as well as ARA) can form potent anti-inflammatory mono-epoxide derivatives (Atone et al. 2019). Furthermore, and relevant to our study, a bimolecular condensation pathway of brain DHA with ethanolamine or a related congener produces a cannabinoidal-like prosurvival adduct, N-docosahexaenoylethanolamide (synaptamide) (Kim and Spector 2018; Watson et al. 2019). Among its known neuronal effects, synaptamide displays anti-inflammatory activity that is mediated to a degree by orphan G-protein coupled receptor 110 (Park et al. 2019).

Results from numerous studies indicate the importance of normal iPLA2 levels and/or activity for proper brain function (Turk et al. 2019), in part reflecting its important role in neuronal membrane phospholipid remodeling that encompasses brain ω–3 DHA availability and turnover. Experiments indicate that normal iPLA2 activity and levels are necessary for synaptic (esp. glutamatergic) stability, plasticity and function (Allyson et al. 2012). Also, genetic iPLA2 VIA (iPLA2β) deficiency in mouse models is associated with reduced brain DHA and its disturbed metabolism (Cheon et al. 2012). In addition, blockade of iPLA2 activity by brain infusion of a selective inhibitor augments rotenone’s deleterious brain mitochondrial effects in a parkinsonian rat model (Chao et al. 2018).

Human mutations of the iPLA2 gene that impair its activity have been linked with infantile-neuroaxonal dystrophy, neurodegeneration with brain iron accumulation, and Karak syndrome (Gregory et al. 2008). Furthermore, the iPLA2 VIA gene is believed to be linked to an autosomal recessive form of Parkinson’s disease (PD), possibly via alpha-synuclein stability dysregulation (Mori et al. 2019). Indeed, a form of regulated necrosis that may underlie PD, ferroptosis, appears to be averted by iPLA2 VIA actions (Sun et al. 2021). Also, in studies with Alzheimer’s subjects, a correlation has been shown between low blood iPLA2 levels and progression from mild cognitive impairment (Gattaz et al. 2014). In contrast, it should be noted that significantly increased iPLA2 can be associated with detrimental brain outcomes as well (Yang et al. 2019). Collectively, results emphasize the importance of normal iPLA2 expression—and regulated mobilization/turnover of DHA—in brain function and neuronal survival. With respect to alcohol, significant depletion of iPLA2 VIA and DHA in the HC, ECX and OB was found in adult rats experiencing highly elevated BACs during the neurotoxic Majchrowicz severe binge procedure (Tajuddin et al. 2014). In concurrent in vitro studies, rat organotypic HC-ECX slice cultures binge-treated with high alcohol (~460 mg/dl) demonstrated iPLA2 and DHA losses accompanying neurodegeneration, and these were inhibited or suppressed by DHA supplementation (Brown et al. 2009; Tajuddin et al. 2014).

Whereas the neuroinflammatory and neurodegenerative mechanisms caused by the high BACs seen in chronic, binge AUD receive ongoing investigations in various laboratories including our own (Kouzoukas et al. 2019; Marshall et al. 2016; Crews et al. 2013), there is little research on potential brain pro-neuroinflammation and its mechanisms promoted by more commonly occurring, relatively moderate BACs from sporadic binge drinking. We report here the effects of exposures of adult male rats to a week of daily upper-moderate BACs on selected neuroinflammatory indicators in the brain regions known to be neurovulnerable to the aforementioned severe binge BACs. Quantitative microglial and unsaturated fatty acid indicators of neuroinflammation were measured in conjunction with a widely-accepted neurodegeneration stain. The results with moderate BACs revealed, in the absence of neurodegeneration, an unexpected selective depletion of hippocampal nonesterified DHA and its anti-inflammatory synaptamide derivative, in association with significant reductions in HC and ECX of iPLA2, the key phospholipase that regulates DHA. Furthermore, daily DHA supplementation counteracted iPLA2 declines in the HC. We surmise that deficits of these metabolically-interrelated anti-inflammatory mediators signal a distinct weakening or diminishment of brain lipid anti-oxidant reserve, and could be a portent of regional neurovulnerability with respect to neurodamage that is caused by more prolonged or higher neurotoxic BACs.

MATERIALS AND METHODS

Animals and alcohol treatment.

All experiments were reviewed and approved by the Loyola University Institutional Animal Care and Use Committee (IACUC) and performed according to the guidelines of the National Institutes of Health. Adult male Sprague-Dawley rats (Envigo, Indianapolis, IN; 9 weeks old, 334 ± 7 g), were handled daily for one week prior to alcohol treatment. They then were intragastrically gavaged once daily (9–10 AM) with alcohol (ethanol, 5 g/kg; 40% in vanilla Ensure Plus) or isocaloric dextrose (for controls) for 7 consecutive days. The midbrain results were obtained from 8–10 week old male C57BL/6 mice (Charles River Laboratories, Wilmington, MA), gavaged once daily with 3 g/kg alcohol (25 % ethanol) or water vehicle for 3 days prior to sacrifice in the M.A. Choudhry laboratory at Loyola, under an IACUC-approved protocol.

DHA supplementation.

DHA (Cat. D2534, Sigma-Aldrich, St. Louis, MO) was dissolved in olive oil (Great Value Classic Olive Oil, Bentonville, AR) to provide a stock concentration of 60 mg/ml. One day prior to the start of alcohol and control binges, 100 mg/kg DHA or olive oil vehicle was delivered by gavage, and then once daily combined with alcohol or control dosing as above for 7 days.

Blood alcohol assays.

Ninety minutes after alcohol gavages on days 3 and 5, 200 μl blood aliquots were collected by tail snip from alcohol-treated rats. Blood was allowed to clot and plasma was taken following rapid centrifugation for 30 min (1942 x g) at 4 °C in small vials that, following kit instructions, were immediately parafilm-covered and lock-capped to prevent alcohol evaporation. BACs were determined using the Pointe Scientific kit (Cat. 23–66-072, Canton, MI). With mice, blood was collected 3 hours after final gavage via cardiac puncture during the euthanasia procedure, and spun down in immediately-capped vials at 8000 rpm for 5 minutes to retrieve plasma for BACs, that were rapidly determined with an AM1 Alcohol Analyzer (Analox Instruments, Stourbridge, UK), per manufacturer instructions.

Brain collection.

Approximately 90 minutes after day 7 gavages, rats were anesthetized with isoflurane and intracardially perfused with ice-cold lactated Ringer’s solution (Cat. 00338–0117-04, Baxter, Deerfield, IL). Following decapitation, brains were removed, halved at the midline, flash-frozen with dry ice/methanol slurry, and stored at −80°C. Similarly, after mice were decapitated 90 min after the 3^rd daily alcohol or water dose, brains were removed and halved at the midline, dissected into discrete regions including the midbrain, flash-frozen and stored at −80 °C prior to analysis.

Cryosectioning and brain tissue staining.

Thawed brains were frozen to optimally-cutting- temperature, and then sagittally-sectioned at 25 μm using a Leica CM3050 S cryostat (Object temperature: −19°C; Chamber temperature: −22 °C). Sections were placed on Diamond-charged glass slides and fixed at room temperature in 4% paraformaldehyde. Staining with Fluoro-Jade B (FJB) (Cat. AG310, Millipore Sigma, Burlington, MA) and the nuclear stain DAPI (Cat. D1306, Invitrogen, Carlsbad, CA) was achieved with sequential (1) 1% NaOH in 80% ethanol (5 min); (2) 70% ethanol (2 min); (3) distilled deionized H₂O (dd H₂O; 2 min); (4) 0.06% KMNO₄ (10 min); (5) dd H₂O (2 min); (6) 0.0004% FJB (with 0.0001% DAPI) (20 min); and (7) dd H₂O (1 min) (x3). Dried slides were cover-slipped using Cytoseal 60 (Cat. 8310-Thermo Scientific, Waltham, MA).

Imaging.

Sections were imaged using semi-automated programming on an Olympus IX80 inverted fluorescent microscope with cellSens software (RRID:SCR_016238 Olympus, Tokyo, Japan). FJB was imaged using a FITC filter, and DAP using a DAPI filter. Images were analyzed using FIJI (RRID:SCR_00285).

Western Blotting.

Brain regions of HC, ECX and OB from rats were microdissected on a dry ice-cooled stage. Protein was harvested in RIPA buffer (Cat. R0278 Millipore-Sigma, Burlington, MA) supplemented with protease and phosphatase inhibitor cocktail (Cat. PPC1010, Millipore-Sigma, Burlington, MA). Mice midbrain samples were treated similarly. Protein concentrations were determined using the bicinchoninic acid assay (Cat. 23225, Pierce BCA kit, Thermo Fisher, Waltham, MA); protein concentrations of 1 μg/μl in 2x loading dye (β-mercaptoethanol) were made, and samples were loaded on 4–20% Express PAGE gels (Cat. M42012, GenScript, Piscataway, NJ) and run at 120 V. Proteins were transferred onto nitrocellulose (0.2 μm) at 70 V for 90 minutes. Blots were blocked for 5 minutes with 1x Rapid Block (Cat. VMRVM325, VWR, Radnor, PA), washed, and then incubated in milk with shaking overnight at 4°C, followed by addition of the primary antibodies: Anti-iPLA2 VIA (1:500) (RRID: AB_11129638, Millipore Sigma, Burlington, MA); Anti-cPLA2 IVA (1:500) (RRID: AB_627288, Santa Cruz Biotechnology, Dallas, TX); Anti-sPLA2 IIA (1:500) (RRID: AB_1527520, BioVendor, Czech Republic); Anti-Iba1 (1:100) (RRID: AB_839506, Wako, Japan); Anti-CD40 (1:500) (RRID: AB_10885918, Bioss, Woburn, MA). Blots were washed in TBST and incubated in horse radish peroxidase-conjugated secondary antibody for 1 hour at room temperature. Anti-GAPDH (1:1000) (RRID: AB_10847862, Santa Cruz Biotechnology, Dallas, TX) was used as loading control. Bands were detected after using SuperSignal West Femto ECL (Cat. 34094, Thermo Fisher, Waltham, MA) substrate and imaged using a BioRad Chemidoc (RRID: SCR_014210, Hercules, VA). Integrated optical densities of blots were determined using FIJI (RRID:SCR_00285).

2.7 │. Lipidomics.

Contents of the free fatty acids (FFAs), ARA, DHA, and the N-acyl-ethanolamide, synaptamide, were quantitated using liquid chromatography/mass spectrometry (LC/MS) as previously described by Smith et.al. (2018), with minor modifications. Brain regions (HC, ECX, OB) were dissected on a dry ice-cooled stage and homogenized using a 7:3 mixture of methanol-water containing 50 μg/mL butylated hydroxytoluene (BHT, Sigma-Aldrich, St. Louis, MO). Homogenate protein levels were determined for each sample using the Pierce™ BCA protein assay kit (Cat 23227, Thermo Fisher Scientific, Waltham, MA). Three deuterated standards, 50 pmol each of d5-DHA and d8-ARA, and 150 fmol d4-synaptamide, were added to a volume of homogenate corresponding to 100 mg protein. Extracts in 7:3 methanol-BHT:water, were loaded onto 30 mg Strata-X polymeric C18 SPE solid phase extraction (SPE) cartridges (Cat. 8B-S100-TAK, Phenomenex, Torrance, CA), washed extensively with water, and eluted with 2.5 mL methanol-BHT. Eluants were dried with a nitrogen stream and reconstituted in a small volume (approximately 30 uL) of methanol-BHT. Samples were injected onto a Zorbax Eclipse Plus C18 column (Cat. 959757–902, Agilent, Santa Clara, CA) and compounds were separated via HPLC using a three-stage tertiary gradient. HPLC solvents were 0.01% acetic acid (Cat. A113–10X1AMP, Thermo Fisher Scientific, Waltham, MA) and consisted of water (A), methanol (B) and acetonitrile (C) (Avantor, Radnor Township, PA). After pre-equilibration of column with A/B (60%/40%), 5 mL extract was injected and the solvent composition was linearly changed to A/B/C (36.3%/15%/48.7%) in 5 min, followed by a linear gradient to A/B/C (13.5%/68.4%/18.1%) over 22 min. HPLC eluant was coupled to a Q-Exactive mass spectrometer (Thermo Fisher, Waltham, MA). Data was collected in the targeted positive MS-MS mode, monitoring the transitions 372.3à62.060 (synaptamide) and 376.3à66.085 (d4-synaptamide). For FFAs data, the mass spectrometer was operated in negative MS scan mode (m/z 290–750). Quantitation of synaptamide was achieved by analyzing the MS-MS peak areas of the characteristic fragments of synaptamide at m/z 62.060 in comparison to that of d4-synaptamide at m/z 66.085. Relevant [M-H]-ion peak areas for FFAs were compared for DHA, ARA, d5-DHA, and d8-ARA at m/z 327.233, 303.233, 332.264, and 311.283, respectively.

Statistics.

One-way ANOVAs determined significant interactions between factors and significant differences between treatment groups where appropriate, and post-hoc analyses consisted of Tukey’s Honest Significant Difference (HSD) test. A repeated measures MANOVA was used to determine differences across days and between groups. Student’s t-tests were used to compare between two groups. Extreme outliers, i.e., those failing Dixon’s Q test, were removed from analyses where noted. Data are presented as mean ± SEM, with statistical difference of p≤0.05 considered significant. All statistical analyses were performed using SPSS Statistics (RRID: SCR_002865, IBM, New York, NY) and graphs were generated using Prism8 (RRID: SCR_002798, GraphPad, San Diego, CA).

RESULTS

Moderate BACs from once-daily binge treatments did not affect rat weight gain and were unaltered by DHA supplementation.

Binge alcohol daily gavages produced average BACs of 155.4 ± 53.89 mg/dl for the alcohol group and 151.1 ± 50.87 mg/dl in the alcohol plus DHA group (Fig. 1B), establishing that DHA supplementation did not significantly alter BACs. While weight loss is typical in severe Majchrowicz binge models, no significant weight changes were observed between the moderate BACs groups and control groups (repeated measure MANOVA, n=5–7, p>0.05) (Fig. 1C). Additionally, there were no significant differences in gavage volumes between all groups on any day, ensuring that groups received relatively equivalent nutrition; the average milliliters for each group across the 7 days were: control 4.66 ± 0.585, alcohol 4.65 ± 0.584, DHA 4.65 ± 0.585, alcohol+DHA 4.67 ± 1.66.

Figure 1: Moderate BACs (~150 mg/dl) in adult male rats from once-daily binges for seven days were not associated with significant weight loss, nor were altered by 100 mg/kg DHA pre-dosing and daily supplementation.

A) Diagram of once-daily binge paradigm. B) BACs from blood after gavage on day 5 of binge alcohol treatment; n=3–4 rats per group. No difference in BACs was apparent when DHA was supplemented by gavage. Student’s t-test, p>0.05. C) Mean weight across each day of treatment. No difference between treatment groups, 8×4 repeated measure ANOVA; main effect of day but not of treatment group (n=5–7; p<0.001; p=0.623), no interaction. D) Weight change during treatment (comparing initial weight on day 1 to final weight on day 8). ANOVA, n=5–7, p>0.05.

Moderate BACs from once-daily binge treatments caused no detectable neurodegeneration in HC and ECX, two brain regions specifically vulnerable to high neurotoxic binge BACs.

To assess possible neurodegeneration, formalin-fixed brain sections of alcohol-gavaged rats were stained with Fluoro-Jade B (FJB), a widely-employed measure of dead or dying neurons. FJB-positive cells were determined to be neurons based on classic neuronal morphology of soma and axons (Schmued and Hopkins 2000; Tajuddin et al. 2014; Kouzoukas et al. 2019). FJB staining was examined in the three key brain regions known to display significant degenerating neurons during the severe Majchrowicz protocol, the ECX, the dentate gyrus of the HC, and OB. The CA3 region of the HC, in which neurodegeneration in the Majchrowicz model is typically minimal, was also examined. In both alcohol-treated and control rats, no FJB-positive labeled cells were evident in the dentate gyrus (Fig. 2A and Fig. 2B), CA3 (Fig. 2C and Fig. 2D), ECX (Fig. 2E and Fig. 2F) and OB (Fig. 2G and Fig. 2H). For positive controls, representative fixed sections of HC from adult male rats previously binge alcohol-gavaged according to the severe Majchrowicz procedure thrice daily for 4 days with up to 9 g alcohol/kg/d (av. peak BACs 90 min after morning gavage, 350–370 mg/dl) (Kouzoukas et al. 2019) were utilized. Following FJB treatment in parallel with brain sections in this study (A through H), sections from the Kouzoukas et al. studies showed dense degenerating dentate granule cells (Fig. 2i, indicated by arrowheads), consistent with published findings (Kouzoukas et al. 2019).

Figure 2: Moderate BACs from once-daily binges for seven days did not cause evident neuronal cell death in dentate gyrus of HC, hippocampal CA3, ECX or OB of adult male rats.

Figures are of FJB staining in brain regions of rats sustaining moderate BACs and control rats. Figures 2A & B show HC dentate gyrus, Figures 2C & D show the cornu ammonis 3 (CA3) region of HC, Figures 2E & F show ECX, and Figures 2G & H show the olfactory bulb. No significant FJB staining was apparent in any of the regions examined. Image I shows a positive FJB staining control with punctate labeling of degenerating neurons (arrows) in fixed section of HC dentate gyrus, obtained from this laboratory’s reported experiments with adult male rats given binges 3x daily in the 4-day Majchrowicz protocol and sustaining BACs >350 mg/dl (Kouzoukas et al. 2019). Images taken at 20x.

Moderate BACs from once-daily binge treatments neither promoted microglial activation nor increased brain pro-neuroinflammatory cPLA2 IVA and sPLA2 IIA levels.

Evidence for neuroinflammatory amplification in the HC and ECX of rats sustaining moderate BACs was sought by examining indicators of microglial activation, and also of increased levels of cPLA2 IVA and sPLA2 IIA. It is established that high BACs from repetitive, severe binge alcohol treatments of the Majchrowicz procedure promote indicators of microglial activation (Marshall et al. 2013; Crews et al. 2013) and increase levels of Ca⁺²-dependent PLA2 isoforms (Tajuddin et al. 2014). Brain samples from rats sustaining moderate BACs along with controls were first analyzed by Western blot for ionized calcium binding adaptor molecule 1 (Iba1). Results showed no significant differences in Iba1 levels between alcohol and control rats in the HC (Fig. 3A) and ECX (Fig. 3B). To verify Iba1 results, levels of cluster of differentiation 40 (CD40), a protein required for microglial activation and upregulated by high BACs and inflammatory cytokines (He and Crews 2008; Marshall et al. 2013), were determined. Data in Fig. 3C and Fig. 3D showed no differences in CD40 expression in HC and ECX, respectively, between rats with moderate BACs and their controls. Regarding neuroinflammatory phospholipid regulation, the effects of moderate BACs on ARA-mobilizing enzymes, cPLA2 IVA and sPLA2 IIA, were then determined with Western blot. Fig. 3E and Fig. 3F show that cPLA2 levels in both the HC and ECX did not significantly differ from control values in samples from rats with moderate BACs. Likewise, sPLA2 levels in these two neurovulnerable regions taken from rats sustaining moderate BACs were not different from sPLA2 levels in controls (Fig. 3G and Fig. 3H).

Figure 3: Moderate BACs in adult male rats from once-daily binges for seven days did not increase microglial-associated neuroinflammatory markers Iba1 and CD40, nor enhance cPLA2 IVA and sPLA2 IIA levels, in HC and ECX.

Graphs show quantification of western blot optical density data from hippocampus (HC) and entorhinal cortex (ECX) for Iba1 (A-B), CD40 (C-D), cPLA2 VIA (E-F), and sPLA2 IIA (G-H) compared to GAPDH levels and normalized to controls between blots. There were no significant differences between alcohol and control results for the microglial markers or enzymes examined. n=3–6/group, T-test, p>0.05.

Moderate BACs from once-daily binge treatments reduced iPLA2 levels in HC and ECX, and supplementation with DHA blocked iPLA2 depletion in HC.

Studies indicate that normal brain iPLA2 levels are fundamental to proper neurofunction. We observed here with Western blotting that moderate BACs over one week markedly depleted iPLA2 VIA in HC (Fig. 4A) and ECX (Fig. 4B). Whereas DHA treatment only did not alter iPLA2 levels in the regions of controls (Fig. 4A and 4B), the supplementation in rats sustaining moderate BACs antagonized iPLA2 depletion and promoted normalization of levels in the HC (Fig. 4A). In ECX, however (Fig. 4B), the greater alcohol-induced iPLA2 decrements (of over 90%, compared to 80% HC deficits) were not significantly counteracted by the DHA doses we utilized.

Figure 4: Moderate BACs in adult male rats from once-daily binge treatments for seven days significantly reduced iPLA2 levels by 80–90% in HC and ECX, with DHA supplementation mitigating iPLA2 deficits in HC.

Graphs show Western blot data quantification of iPLA2 VIA relative to GAPDH levels and normalized to controls between blots. Significant reductions of iPLA2 were determined in the HC (Figure 4A; ~80%), and ECX (Figure 4B; ~90%). Daily doses with 100 mg/kg DHA as described completely prevented the depletion of iPLA2 levels in HC (A), but was insufficient to counteract the extensive iPLA2 decline in ECX (B). ANOVA, Tukey’s, n=5–7, ns=not significant, *p<0.05, ***p<0.001.

Moderate BACs from once-daily binge treatments significantly decreased HC levels of nonesterified DHA and synaptamide, without altering levels of nonesterified ARA.

Since iPLA2 is the primary phospholipase selectively liberating DHA from glycerophospholipids, it was reasoned that lowered iPLA2 might be associated with reduced nonesterified DHA. Lipidomic analyses of free fatty acids and metabolites were performed on HC and ECX samples from rats sustaining moderate BACs, concomitant with control samples. Nonesterified DHA was significantly reduced by moderate BACs in the HC, but failed to reach significance in the ECX (Fig. 5A and Fig. 5B, respectively). Furthermore, DHA losses in HC were accompanied by significant decreases of synaptamide in this brain region (Fig. 5C). Consistent with unchanged DHA levels in ECX of rats sustaining moderate BACs, synaptamide was not decreased in this region (Fig. 5D). Measurements of ARA levels in HC and ECX showed that this omega-6 fatty acid was unaltered from control levels by the moderate binge alcohol treatment (Fig. 5E and Fig. 5F), which correlates with the unchanged levels of cPLA2 and sPLA2 (Figure 3.3), phospholipases considered principally responsible for ARA mobilization.

Figure 5: Moderate BACs in adult male rats from once-daily binges for seven days significantly reduced HC levels of nonesterified DHA and DHA-derived synaptamide, while not altering ARA levels in HC and ECX.

Figure 5A, 5C, & 5E display results of lipidomics analysis of the HC, and Figure 5B, 5D, & 5F show results of lipidomics analysis of the ECX taken from rats sustaining moderate BACs (alcohol) and control rats. DHA and synaptamide in alcohol rats were significantly reduced ~33% below control levels in HC, but were unchanged from control levels in the ECX. Levels of ARA were also unchanged in alcohol-treated rats relative to controls in both HC and ECX. Amounts were quantified by HPLC-MS analysis of area under the curve compared to known standards according to methods described (Smith et al. 2018). n=4–5, Two-tailed T-test, *p<0.05.

Moderate BACs from once-daily binge treatments over several days in adult male mice significantly depleted midbrain iPLA2 levels.

Midbrain regions taken from adult mice gavaged once daily for 3 consecutive days with either alcohol/water or water only, as described in Methods, were provided by the Choudhry laboratory at Loyola University Chicago and analyzed for iPLA2 VIA protein levels. Western blot data revealed significantly lower iPLA2 levels in midbrains of alcohol-treated mice compared to water-treated controls (Fig. 6B). For reference, the midbrain area as diagrammed in Fig. 6C encompasses HC and ECX.

Figure 6: Moderate BACs in adult male mice from once-daily binges for three days significantly reduced midbrain iPLA2 levels.

A) Experimental design: Adult mice gavaged with 25% alcohol (3 g/kg ethanol in distilled deionized water) or vehicle (water) once daily, with sacrifice by decapitation 180 min after 3^rd alcohol dose. B) Midbrain tissues from alcohol-gavaged mice had significantly reduced iPLA2 levels compared to iPLA2 levels in water-gavaged control mice. n=7 per group, T-test, *p<0.05. C) Diagram of midbrain brain region examined between dashed red lines that includes HC and ECX. Note: BACs, 180 mg/dl from sacrifice blood taken on day 3, were estimated to have reached a peak level of ~200 mg/dl at 90 min.

DISCUSSION

Our reported studies in adult rats as well as adult-age organotypic brain slice cultures have shown that high circulating BACs or alcohol concentrations in culture of ~460 mg/dl (100 mM) significantly elevate levels and/or activity of cPLA2 IVA and sPLA2 IIA, and increase pro-inflammatory ARA, which is primarily mobilized by these phospholipases, while reducing levels of iPLA2 VIA and nonesterified anti-inflammatory DHA, which is regulated by this PLA2 (Tajuddin et al. 2014). The findings here are that upper-moderate but nontoxic BACs still suppress levels of iPLA2 and DHA—as well as the novel DHA-derived synaptamide—while not apparently involving other prominent neuroinflammatory mediators or pathways. We suggest that the reductions may have an early pro-active role in compromising the anti-inflammatory capacity or reserve of brain regions that are well known to be neurovulnerable to much more severe binge BACs. Notably, the binge alcohol model in this report produced BACs that are seen in a much larger demographic group than in severe, chronic AUD. Also, since there were no weight differences due to alcohol treatment, it is reasonable to attribute changes from alcohol itself and not dietary deficiencies, as seen in other AUD-associated neuropathologies such as Wernicke-Korsakoff encephalopathy.

Average BACs in the study were less than originally predicted, but as alcohol supplements were prepared in vanilla Ensure Plus and supplemented with olive oil, the vehicle for DHA supplementation, the lower-than-expected BAC is attributed in part to the high caloric intake during gavage. It is well-known that eating while consuming alcohol changes alcohol absorption, and thus peak BACs (Ramchandani et al. 2001). The olive oil vehicle in combination with the Ensure Plus possibly affected absorption to produce lower-than-expected peak BACs. The alcohol assay itself might be contributing to lower-than-expected levels due to evaporation during centrifugation, but we consider this unlikely due to the presence of control alcohol samples,

As described, there was no FJB staining (neurodegeneration) in the examined brain regions of rats exposed to high-moderate peak BACs of ~150 mg/dl. The lack of FJB staining indicates that neurodegeneration is not occurring over a week timespan at these blood alcohol levels, but we might expect that pervasive neuroinflammatory changes and even neurodamage would result if the experiments were greatly extended at these BACs. Overall, the results suggest that, at least semi-chronically, BACs in high-moderate ranges might promote important initial omega-3 fatty acid-related deficits that could foreshadow a neurovulnerable brain state, thus priming susceptible regions for overt neurodamage and widespread neuroinflammation at much higher BACs.

The mechanisms underlying moderate alcohol’s apparently selective effect on iPLA2 VIA levels are not experimentally addressed here. However, as iPLA2 is largely responsible for catalyzing DHA mobilization from membrane phospholipids, it follows that there should be less nonesterified DHA and its derivatives/metabolites to act via downstream signaling pathways, and this was observed in HC. Since DHA and its metabolites are principally anti-inflammatory, it is reasonable that anti-oxidative neuroresilience in memory-related brain regions such as the HC is likely to be diminished. However, while nonesterified DHA reductions by the moderate BACs are associated with, and presumably coupled to, iPLA2 depletion in the HC, they are not with the iPLA2 deficit in ECX. The HC underscores how iPLA2 might play a key role in HC in mediating the pool of nonesterified DHA, but the ECX requires further study. Of potential significance, decreased HC synaptamide levels mirrored the reduction in DHA, consistent with synaptamide being a direct metabolite of this important brain omega-3 fatty acid. Given synaptamide’s known neurobeneficial effects (Kim and Spector 2018), its depletion from moderate BACs implies that it may have a key anti-inflammatory, neurosurvival role in this memory region. It is also possible that other DHA derivatives reported to have roles in DHA’s anti-inflammatory and pro-resolving effects, such as NPD1 and mono-epoxylated DHA metabolites (Bazan 2018; Atone et al. 2019), might also be similarly reduced by moderate BACs.

Moderate BACs caused no cPLA2 and sPLA2 elevations in the HC or ECX, so the lack of change in nonesterified ARA levels is reasonable. Furthermore, the absence of increases in microglial activation markers provides indication that frequently-associated glial-directed pro-inflammatory players or events (inflammasomal activation, pro-inflammatory cytokines and others) are not elicited by short-term exposure to these high-moderate BACs. Overall, the findings indicate that a potentially selective neuroinflammatory outcome from moderate BACs is a diminution in lipid anti-inflammatory effectiveness selectively involving iPLA2, DHA and DHA-derived synaptamide.

Providing DHA beginning a day before and then throughout a week’s exposure to moderate BACs was sufficient to prevent the alcohol-induced iPLA2 deficits in HC (and OB, not shown). How supplemented DHA acts to counter depletions and normalize iPLA2 levels at a molecular and cellular level is unknown, but is possible that the replenished DHA or its active metabolites—notably, synaptamide—directly inhibit alcohol’s basic molecular mechanism(s)—which, as previously mentioned, also are unclarified. Because of the homeostatic importance in brain of normal iPLA2 levels and activity, our finding is consistent overall with many animal and brain cell studies in which supplemention with omega-3 fatty acids or DHA alone has significant neurobeneficial actions and effects. As indicated in reviews summarizing studies of DHA supplementation (Salem et al. 2015), with and without PLA2 inhibitors (Farooqui et al. 2006), such studies based on specific brain insults or deficits often have been mainly neurobehavioral, such as improved cognitive measures or antagonism of experimental depression, although some do include anti-inflammatory or prosurvival effects. Particularly germane to our findings are experiments with adult rats subjected to concussive brain injury (Wu et al. 2011) that show depletion of iPLA2 VIA in concert with brain-derived neurotrophic factor (BDNF), and that dietary DHA supplementation inhibited the traumatic injury by normalizing BDNF and iPLA2 levels. This brain trauma study suggests important links between brain BDNF, iPLA2 and DHA that may relate to alcohol; indeed, high binge BACs, with associated withdrawal episodes, are a form of brain trauma (K. Sripathirathan et al. 2009).

Although experiments of the brain trio are few, other studies have made BDNF-DHA connections. For example, reduced frontal cortical BDNF expression was observed after 15 weeks of dietary deprivation of DHA and other ω–3 fatty acids in rats, and DHA supplementation of rat cortical astrocyte cultures promoted BDNF protein induction (Rao et al. 2007). Additionally, adult rats treated with clozapine, a treatment for schizophrenia and bipolar disorder, showed upregulation of both iPLA2 and BDNF messages and protein (Kim et al. 2012). Furthermore, in adult animal models, repetitive or semi-chronic alcohol exposures (sometimes at unspecified BACs) have been linked to brain BDNF reductions (Davis 2008; Xu et al. 2015; Palmisano and Pandey 2017; Motaghinejad et al. 2020). Consequently, a reasonable hypothesis to be tested in our semi-chronic binge model emerges: moderate BACs trigger depletion of both brain BDNF and linked iPLA2 levels that are thwarted by DHA supplementation, such that BDNF might thus be a key DHA-responsive modulator of molecular events facilitating iPLA2 normalization.

As mentioned, unlike in the HC with ca. 80% iPLA2 reduction, the daily DHA supplementation we employed could be insufficient to counteract alcohol-induced iPLA2 decrements in ECX. Experiments are needed to determine whether the greater iPLA2 depletion (over 90%) requires a higher DHA dose to normalize iPLA2 in this neurovulnerable region, or whether there are other competing or antagonistic concerns. Also to be noted in Figure 6 are midbrain iPLA2 reductions in mice binge-exposed to moderate BACs, implying an interspecies consistency of brain susceptibility of iPLA2 to alcohol. A closing point that acknowledges important growing attention to adolescent binge drinking and its effects on brain development (Jones et al. 2018; Morris et al. 2018) is the potential value in examining brain regional iPLA2, nonesterified DHA and synaptamide in adolescent animals subjected to moderate BACs. Such studies could afford important insights into possible diminished regional anti-inflammatory lipid neuroreserves during brain development.

Abbreviations:

ARA

arachidonic acid

AUD

alcohol use disorder

BACs

blood alcohol concentrations

BHT

butylated hydroxy toluene

DG

dentate granule

DHA

docosahexaenoic acid

ECX

entorhinal cortex

FJB

Fluoro-Jade B

HC

hippocampus

NPD1

neuroprotectin-1

OB

olfactory bulb

PAGE

polyacrylamide gel electrophoresis

PLA2

phospholipase A2