Female wild-type c57BL/6J mice (Jackson Laboratories, Bar Harbor, ME) weighing ≥ 19 g and ≥ 10 weeks of age at the beginning of the study were individually housed in cages with shredded filter paper and wooden blocks for enrichment and to prevent malocclusion from receiving a liquid-only diet. Mice were acclimated for 2 weeks in temperature (22°C), light, and humidity-controlled (40–60%) conditions with a 12 h light/dark cycle. During acclimation weeks, mice were given free access to the liquid Lieber DeCarli diet with no ethanol (Bio-Serv, Flemington, NJ, USA) following the model described by Bertola et al. (34), with modifications. After the acclimation period, mice were randomly assigned to groups (where feed was given ad libitum): (1) No-EtOH (n = 12); (2) EtOH [(n = 12) 5.5% (v/v)]-containing LDC diet; and (3) DHM (n = 12; 6 mg/mL) + EtOH-containing LDC diet, for a total of 5 weeks. Every morning, fluid intake was recorded by measuring the meniscus on the graduated feed tube. Mice in the DHM group were exposed to ethanol-only for 2 weeks prior to DHM supplementation, which lasted for the remainder of the study; the feeding paradigm was isocaloric between groups. DHM, [(2R, 3R)-3, 5, 7-trihydroxy-2-(3, 4, 5-trihydroxyphenyl)- 2,3-dihydrochromen-4-one], HPLC grade, >98%, MW 320.25 was purchased from Master Herbs Inc., Pomona, CA. The LDC diet is a robust forced-drinking model that induces severe liver disease with a potential for a high mortality rate. Therefore, for this proof-of-concept study, a single dose of 6 mg/mL was used as a comparison to the 10 mg/kg dose delivered via i.p. in our previous publications. After the study period ended, mice were euthanized via CO₂ exposure followed by cardiac puncture. Blood was collected and kept at room temperature for 45 min and serum was separated by centrifugation for 10 min at 10,000 × g in 4°C and stored at −80°C until use; livers were harvested and frozen in nitrogen-isopentane and stored at −80°C until use or fixed in 10% formalin and embedded in paraffin. Animals used in the study were considered and handled in adherence to the University of Southern California’s Department of Animal Resources Institutional Animal Care and Use Committee (IACUC) policies and guidelines.

Lipid droplets were stained using Oil Red O Staining Kit (Lifeline Cell Technology, San Diego, CA, USA) on frozen liver sections (10 μm thick). Liver sections were also stained with Hematoxylin and Eosin (H&E) staining kit (Abcam, Boston, MA, USA). Antibodies against p62/SQSTM-1 (1:400, Cell Signaling Technology, Danvers, MA, USA); LC3B (1:1,000, Cell Signaling Technology, Danvers, MA, USA); PLIN-1 (1:200, Cell Signaling Technology, Danvers, MA, USA); anti-CD68 (1:250, Cell Signaling Technology, Danvers, MA, USA); and Alexa Fluor 405, 488, and 647 secondary antibodies (1:250, Cell Signaling Technology, Danvers, MA, USA) were used for visualization. Images were acquired using Cytation 5 Cell Imaging Multi-Mode Reader (BioTek, Winooski, VT, USA) and Zeiss LSM880 w/Airyscan Confocal Laser Scanning Microscope (Carl Zeiss Microscopy, White Plains, NY, USA), and were analyzed using ImageJ software (ImageJ; Coloc2 Fiji software) and Zen (Black and Blue versions) imaging analysis software (Carl Zeiss Microscopy, White Plains, NY, USA). Lipid droplet density and size were analyzed using whole image analysis on the ImageJ Color Threshold software.

Protein expression (Atg7, PLIN-1, p62, CETP, and LCAT) was analyzed using protein extracts (60–125 mg for immunoprecipitation) from liver homogenates that were isolated using Dynabeads Magnetic Beads (Thermo-Fisher Scientific, MA, USA), and visualized via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), where bands were detected by chemiluminescent reaction. Signal density was quantified by densitometry using ImageJ software: Atg7, PLIN-1, p62/SQSTM-1, CETP, and LCAT were analyzed using immunoprecipitation, while LC3B and Beta-actin antibodies purchased from Cell Signaling Technology, CA were analyzed from Western blots (diluted 1:1,000, while 10 μg of antibody was used for IP).

The following assays were measured from serum: aspartate and alanine aminotransferase (AST and ALT) levels were measured using AST and ALT activity assays (Sigma Aldrich, St. Louis, MO, USA). Free cholesterol and cholesteryl esters were measured using the Cholesterol Fluorometric Assay Kit (Cayman Chemical, Ann Arbor, MI, USA). LDL/VLDL levels were measured using the Cholesterol Assay Kit (Abcam, Boston, MA). Circulating triglyceride levels were measured from liver homogenates (∼100 mg) and serum using Triglyceride Colorimetric Assay Kit (Cayman Chemical, Ann Arbor, MI, USA).

Mitochondrial oxidative phosphorylation system (OXPHOS) in complexes I, II, and IV were analyzed using Complex I Enzyme Activity Colorimetric Assay Kit, Complex II Enzyme Activity Microplate Assay Kit, and the Complex IV Rodent Enzyme Activity Microplate Assay Kit (Abcam, Boston, MA, USA) using isolated mitochondria that were purified from the liver tissues using the Mitochondria Isolation Kit for Tissue (Abcam, Boston, MA, USA).

Cytokine levels were measured using Proteome Profiler Array Mouse Cytokine Array Kit Panel A (R&D Systems, Minneapolis, MN, USA) and signal density was quantified by densitometry using ImageJ software.

Immunohistochemistry images were analyzed using n = 3–4 from each group and 4 different sections were analyzed per sample. Biochemical assays were conducted using 3–4 samples from each group. Data are presented as means ± standard deviation. Statistical analysis included one-way ANOVA along with Bonferroni multiple comparison tests using Prism 9.3 (GraphPad Software, Inc., San Diego, CA, USA), where p ≤ 0.05 was considered statistically significant.

To investigate the utility of oral DHM, mice in the DHM group received ethanol-only treatment for 2 weeks prior to DHM supplementation to assure the initiation and development of ALD pathology. A hallmark of early ALD is hepatic steatosis, characterized by the accumulation of LDs throughout the liver and disruptions in lipid homeostatic conditions (35–37). H&E and Oil Red O-staining of liver tissue sections demonstrated increased steatosis in the EtOH group which was alleviated by DHM treatment (Figures 1A, B; scale bars 200 μm). LDs are synthesized by nearly all cells, and size varies considerably among different cell types in response to environmental cues, particularly in the liver. Chronic ethanol consumption alters hepatocyte LD properties, including increased size and cellular distribution (38). Interestingly, mice in the DHM group exhibited a wide range of lipid accumulation and distribution in addition to significantly larger LDs, compared to all groups. In a previous study, it was shown that heterogeneous lipid distribution within the hepatocyte population, similar to what is observed in the DHM group, is a potential hepatoprotective social organization that reduces lipotoxicity within the overall region, compartmentalizing lipids within a cell population (39). The authors also reported that LD heterogeneity is not only reversible and variable (depending on intracellular-environmental factors) but also allows for the reduction of lipotoxicity between cells by exchanging LD content over time. We found a wide range of noticeable hepatocyte-to-hepatocyte LD heterogeneity, which was more prominent in the group receiving DHM. LD size was also found to be varied across groups, with the mean LD sizes in the No-EtOH group measuring at 3.64 μm², EtOH-only at 7.37 μm², and DHM measuring at 9.88 μm². Mice receiving EtOH had larger LDs than the No-EtOH group (^#0.0063; Figure 1C), and the difference was even greater in the mice fed DHM (^##< 0.0001; Figure 1C).

As a measure of overall hepatic health and function following DHM administration, we next analyzed the levels of circulating aspartate and alanine aminotransferases (AST and ALT). As shown in Figure 2A, there was a significant decrease in AST levels in mice receiving oral DHM compared to mice in the EtOH-only group (^Ψ0.029, ^ΨΨ0.0006, ^ΨΨΨ0.024); Figure 2B shows significantly lowered levels of ALT in mice receiving oral DHM when compared to the EtOH-only fed mice (*0.027, **0.0014, and ***0.0046). The levels of circulating lipids were measured and show that DHM administration significantly lowers total free cholesterol levels, (^#0.0097, ^##0.0227; Figure 2C). Although not significant, the levels of LDL/VLDL were reduced with DHM administration, like those measured in the No-EtOH group (Figure 2D). When interpreting the results obtained from this study, it is important to note that the LDC diet is considered as a high fat diet, where 35% of calories are derived from FFAs: 23.5 g/L of monounsaturated and 5.2 g/L unsaturated fats. The No-EtOH group is isocaloric to the other groups and therefore is also receiving the high fat diet, which may influence the amount and types of lipids in circulation compared to the EtOH-receiving groups.

Lipid droplets are primarily composed of triglycerides (TGs) and cholesteryl esters (CE) (37). Although TGs are not considered determinants of lipotoxicity (6), conversion of free fatty acids (FFAs) into TGs, as well as FFA utilization in CEs via esterification (40), essentially acts to neutralize the reactivity of and damage caused by excessive FFAs (41, 42). Accordingly, cholesteryl esters and hepatic TG levels were found to be increased in the group receiving DHM (Figures 2E, F), while circulating TG levels were normalized by DHM to levels similar to those measured in the No-EtOH group (*0.022; Figure 2G). Two facilitators of cholesterol exchange and transport are cholesteryl ester transfer protein (CETP) and lecithin-cholesterol acyltransferase (LCAT). CETP is a known mediator in the transfer of cholesteryl esters from HDL to LDL/VLDL (43). LCAT is a key enzyme involved in the esterification of free cholesterol into cholesteryl esters and facilitates the metabolism of cholesterol (44). As demonstrated in Figure 2H, we found that CETP expression was increased in the EtOH-only group compared to the DHM group, and found that LCAT expression was highest in the DHM fed group.

Lipid droplet membranes are coated with various components, including lipid droplet-associated proteins belonging to the perilipin family (PLIN-1-5) that assist in the regulation of LD synthesis and cytosolic lipase activity. Lipophagy involves the recruitment of selective autophagy proteins such as sequestosome-1 (p62/SQSTM-1), microtubule-associated protein 1 light chain 3 beta (LC3B), and PLIN-1, which when combined are recognized as defense mechanisms against oxidative stress (45–48). Ethanol is known to trigger the selective interactions of p62/SQSTM-1, LC3B, and PLIN-1 using in vitro models during LD clearance (45). To assess the effect of DHM on ethanol-induced interactions between selective autophagy-associated proteins, we analyzed the presence and interactions between p62/SQSTM-1; LC3B; and PLIN-1. We began by confirming the interaction between p62/SQSTM-1 and PLIN-1 using liver sections that were stained for p62/SQSTM-1 (green), PLIN-1 (red), and nuclei (blue). As illustrated in Figure 3 (scale bars 5 μm), we confirmed the presence of and interactions between p62/SQSTM-1 and PLIN-1 in all three groups as shown from their colocalization. Mice receiving ethanol had noticeably higher levels of p62/SQSTM-1 + PLIN-1 interactions than the No-EtOH group (Figure 4B). DHM-fed mice had even greater colocalization events between PLIN-1 and p62/SQSTM-1, where the interactions were widely distributed across LD surfaces, compared to the EtOH-only group (Figure 3 inset).

Autophagy is a diverse mechanism that follows several pathways based on cellular demands. There are over 32 different autophagy-related genes (Atg) that activate the formation of double-membrane structures that deliver cytoplasmic components to lysosomes for degradation. Atg7 is a ligase that has ubiquitin E1-like activity which facilitates interactions and complexations between other autophagy-related genes. These subsequently interact with other Atg proteins, forming a much larger complex that binds to LC3B, a molecule that is essential for autophagosome structure, formation, and cargo recognition (47, 49). LC3B interacts with cargo adaptor protein (p62/SQSTM-1) that binds to poly-ubiquitinated cargo, and is a classical selective autophagy receptor (48).

Selective autophagy occurs when p62/SQSTM-1 and LC3B interact (50), resulting in the formation of an autolysosome that is directed to ubiquitinated PLIN-1 proteins found on LDs, resulting in lipophagy activity. The colocalization of p62/SQSTM-1 + PLIN-1 + LC3B is illustrated in Figure 4A (upper and lower magnified images scale bars are 10 and 5 μm, respectively). Additionally, we quantified levels of interactions by measuring the correlation between two proteins using Pearson’s Correlation Coefficient (PCC) analysis, where values closer to 1.0 confirm the strength of correlation. As shown, interactions between PLIN-1 + p62/SQSTM-1 were significantly higher in mice receiving DHM with a mean PCC of 0.344 (*0.04; brown puncta), compared to the EtOH-only group with a mean PCC of 0.154 (Figure 4B). Our results also show that mice in the group receiving EtOH-only demonstrated an increased colocalization of p62/SQSTM-1 + LC3B (^## 0.008; cyan puncta) when compared to the No-EtOH group; the increase was more apparent when comparing the No-EtOH group to mice fed DHM (^#< 0.0001) (Figure 4C). Our data show that DHM-fed mice had a greater and wider range of colocalization events and activity. Quantification (via immunoprecipitation) of the expression levels of Atg7, PLIN-1, p62/SQSTM-1 were analyzed (Figure 4D), and LC3B: β-Actin (via Western blots) are shown in Figure 4E, supporting the findings from histological colocalization analyses.

Chronic ethanol consumption leads to loss of mitochondrial function and increased production of reactive oxygen species (ROS), promoting oxidative stress, particularly in the mitochondria. Damage, brought on by increases in ROS to mitochondrial proteins and DNA, decreases mitochondrial function due to the breakdown of these complexes (51). Lipid metabolism takes place in the mitochondria, where fatty acids undergo β-oxidation. We measured mitochondrial health by analyzing the activity of complexes I, II, and IV from isolated mitochondria. Our results show that DHM had a significant effect on restoring complex II activity (*0.03; Figure 5B). Although the changes were not significant, DHM led to increases in complex I activity (Figure 5A), while normalizing the activity of complex IV (Figure 5C).

Next, we investigated the effects of DHM on ethanol-induced inflammation. The livers of mice were stained with CD68, a biomarker for immune cells of the monocyte lineage, such as monocytes and macrophages. As illustrated in Figure 6, mice in the EtOH-only group had larger bursts of monocyte/macrophage infiltration clouds (green puncta) when compared to the No-EtOH and DHM-fed mice. These observations were further confirmed by conducting a cytokine panel assay that measured several different circulating pro-inflammatory cytokines from serum (Supplementary Figure 1). We found that oral DHM administration reduced circulating levels of pro-inflammatory cytokines and immune cell chemokines that are traditionally associated with ethanol-induced inflammation. TNF-α promotes acute inflammation and is one of the critical inflammatory cytokines in ALD progression and liver injury, as it contributes to the production of other pro-inflammatory cytokines (33). DHM supplementation reversed the significant elevations in TNF- α levels in mice receiving ethanol-only as demonstrated in Figure 7A (*, **, and ***< 0.0001), while also normalizing levels of IFN-γ (^#< 0.0001, ^##0.0018, and ^##0.001) closer to those of the No-EtOH group (Figure 7B). DHM receiving mice had reduced levels of IL-1β (^Ψ< 0.0001, ^ΨΨ0.027, and ^ΨΨΨ0.0007) compared to the EtOH-only mice, with levels close to the No-EtOH group (Figure 7C). Although IL-1β is not produced in a healthy liver, it is secreted by activated inflammasomes during excessive alcohol consumption and is an essential cytokine in giving rise to Th17 cells that subsequently secrete IL-17 (52, 53).

Mice receiving EtOH-only had significantly higher levels of circulating IL-17 overall, which was ameliorated with DHM supplementation (Figure 7D; ^ꝉ< 0.0001, ^ꝉꝉ0.003, and ^ꝉꝉꝉ< 0.0001). IL-17 is a potent pro-inflammatory cytokine that has received much attention for its synergistic effects with other inflammation promoting cytokines during ALD pathogenesis that were also reduced in the DHM-fed group, such as IL-6 (Figure 7E; *0.007), IL-1β, and IL-1α (Figure 7F; ^ꝉ< 0.0001, ^ꝉꝉ0.047, and ^ꝉꝉꝉ< 0.0001) (52–54). IL-17 induces the expression of hematopoietic cytokines and chemokines such as granulocyte-macrophage-colony stimulating factor (Figure 8A: GM-CSF; ^#< 0.0001, ^## 0.003, and ^###0.0001), macrophage-colony stimulating factor (Figure 8B: M-CSF; ^{ꝉ,ꝉꝉ, ꝉꝉꝉ}< 0.0001), granulocyte-colony stimulating factor (Figure 8C: G-CSF; *< 0.0001, **0.005, and ***0.0004), neutrophil activating and chemotactic chemokine, CXCL1 (Figure 8D; ^#< 0.0001, ^##0.0001, and ^###< 0.0001), and B-cell recruiting CXCL13 (Figure 8E; *0.023, and **0.002) (53, 55–58). DHM-fed mice had lower levels of IL-3 (Figure 8F; ^#0.035), a cytokine that amplifies acute inflammation (59) and works in coordination with GM-CSF to promote pathogenic clearance during chronic inflammation (60). Additionally, CXCL2/MIP-2 (macrophage inflammatory protein-2), is synthesized by a variety of immune cells to recruit neutrophils in response to damage and acute liver injury (61). CXCL2 was significantly increased in mice receiving EtOH-only, and that increase was reversed with DHM supplementation (Figure 8G; ^#,##< 0.0001).

Mice receiving DHM supplementation had increased anti-inflammatory cytokine levels compared to mice fed EtOH-only. IL-1ra is a receptor antagonist to members of the IL-1 family of pro-inflammatory cytokines, and has neutralizing and protective effects against IL-1 activity (62). Mice in the DHM group had levels of IL-1ra that were nearly identical to the No-EtOH group and significantly lower than the EtOH-only group (Figure 9A; *, **< 0.0001). Following excessive ethanol intake and burn injury, IL-27 has been shown to promote liver regeneration by enhancing liver progenitor cell expansion and differentiation as well as intestinal barrier repair following ethanol intoxication (63, 64). In our study, as shown, DHM-fed mice had significantly higher levels of IL-27 in circulation than that of the No-EtOH group (Figure 9B; *0.034) and nearly twice as much as those in the EtOH-only group.