Coordinated action of a gut-liver pathway drives alcohol detoxification and consumption

Yaojie Fu; Bryan Mackowiak; Yuhong Lin; Luca Maccioni; Taylor Lehner; Hongna Pan; Yukun Guan; Grzegorz Godlewski; Hongkun Lu; Cheng Chen; Shoupeng Wei; Dechun Feng; Janos Paloczi; Huiping Zhou; Pal Pacher; Li Zhang; George Kunos; Bin Gao

doi:10.1038/s42255-024-01063-2

. Author manuscript; available in PMC: 2025 Sep 28.

Published in final edited form as: Nat Metab. 2024 Jun 20;6(7):1380–1396. doi: 10.1038/s42255-024-01063-2

Coordinated action of a gut-liver pathway drives alcohol detoxification and consumption

Yaojie Fu ^1,⁶, Bryan Mackowiak ^1,⁶, Yuhong Lin ¹, Luca Maccioni ^1,², Taylor Lehner ¹, Hongna Pan ¹, Yukun Guan ¹, Grzegorz Godlewski ², Hongkun Lu ¹, Cheng Chen ^1,², Shoupeng Wei ³, Dechun Feng ¹, Janos Paloczi ⁴, Huiping Zhou ⁵, Pal Pacher ⁴, Li Zhang ³, George Kunos ², Bin Gao ¹

PMCID: PMC12475873 NIHMSID: NIHMS2104873 PMID: 38902331

Abstract

Alcohol use disorder (AUD) affects millions of people worldwide, causing extensive morbidity and mortality with limited pharmacological treatments. The liver is considered as the principal site for the detoxification of ethanol metabolite, acetaldehyde (AcH), by aldehyde dehydrogenase 2 (ALDH2) and as a target for AUD treatment, however, our recent data suggest that the liver only plays a partial role in clearing systemic AcH. Here we show that a liver-gut axis, rather than liver alone, synergistically drives systemic AcH clearance and voluntary alcohol drinking. Mechanistically, we find that after ethanol intake, a significant proportion of AcH generated in the liver is excreted via the bile into the gastrointestinal tract where AcH is further metabolized by gut ALDH2. Modulating bile flow significantly affects serum AcH level and drinking behavior. Thus, combined targeting of liver and gut ALDH2, and manipulation of bile flow and secretion are potential therapeutic strategies to treat AUD.

Introduction

The remarkable role of the liver in alcohol metabolism has been well known for decades¹. Breakdown of alcohol in liver is mainly processed by a two-step enzymatic reaction. Alcohol dehydrogenase (ADH) is responsible for transforming ethanol into a toxic intermediate byproduct, acetaldehyde (AcH). Then, aldehyde dehydrogenase ALDH, mainly ALDH2, converts AcH to the less toxic compound acetate for further elimination. The importance of AcH in the regulation alcohol consumption was first recognized via the alcohol-induced “flushing” reaction common in the east Asian population^2,3. This reaction is due to a polymorphism in ALDH2 that reduces ALDH2 activity by ~90% in heterozygote or ~99% in homozygote carriers, causing AcH to accumulate after ethanol consumption and leading to an aversive phenotype⁴. Accordingly, people who have the inactive ALDH2*2 polymorphism are protected from developing alcohol use disorder (AUD), which has led to ALDH2 inhibition being utilized as a therapeutic strategy for treating AUD^5,6. Non-specific ALDH2 inhibitor disulfiram is an FDA approved treatment for AUD, but its use is limited due to off-target effects and lack of efficacy in certain populations^7–9. Second generation ALDH2 inhibitors are in development, but it is not clear what organs should be targeted for ALDH2 inhibition and whether central ALDH2 inhibition is required for efficacy¹⁰. Therefore, a better understanding of ALDH2-mediated AcH clearance may lead to the development of better therapeutics for AUD.

While increasing the amount of ethanol-derived AcH may be beneficial to reduce ethanol consumption, AcH is also toxic and is classified as a carcinogen¹¹. In fact, ethanol consumption and the ALDH2*2 polymorphism exert synergistic increases in risk for development of esophageal, head and neck, and liver cancer, as well as various forms of cardiovascular disease^12–14. With ethanol consumption increasing among people heterozygous for the ALDH2*2 polymorphism over the past 20 years, it is more important than ever to develop a comprehensive understanding of ALDH2-mediated AcH clearance in different organs to accurately assess the risks for organ-specific cancer development^12,15.

Early publications suggest about 90% of ethanol and 95% of acetaldehyde are metabolized in liver after alcohol consumption^16,17. However, this conventional notion was strongly challenged by our recent study demonstrating that deletion of liver Aldh2 only partially reduces blood AcH clearance ~30% compared to global Aldh2 deletion¹⁸. This raised a fundamental question – how is the majority of circulating AcH cleared from the body if liver is not the exclusive site? Despite wide expression of Aldh2 in many organs, single organ ALDH2 knockout (KO) in gut epithelial cells or myeloid cells exhibited no increase in serum AcH while adipose tissue ALDH2 knockout only exhibited a small increase in serum AcH. In addition, ALDH2 knockout in forebrain neurons and glial cells did not alter drinking behavior^18,19. Thus, mechanisms of systemic AcH clearance and the physiological significance of other organs in ALDH2-mediated AcH metabolism remain elusive.

Here, using several genetic mouse models, surgical approaches, and in vivo compound administration, we demonstrate that cooperative action of gut and liver, rather than liver alone, is mainly responsible for systemic AcH clearance. In addition, bile efflux of AcH from liver to gut is a critical pathway mediating AcH excretion and turnover. Modification of bile secretion or targeting ALDH2 in gut and liver significantly affect alcohol detoxification and drinking preference. Our studies provide a physiological basis for rethinking alcohol metabolism, AcH-mediated cancer risk, and a novel perspective for clinical treatment of AUD.

Results

Liver and gut ALDH2 synergistically clear blood AcH

Our previous study revealed that the liver ALDH2 only accounts for ∼30% blood AcH clearance, deletion of the Aldh2 in adipocytes slightly reduced blood AcH clearance, and deletion of the Aldh2 in myeloid cells or gut epithelial cells did not affect blood AcH clearance¹⁸. To search for other organs that contribute to systemic AcH clearance, we generated numerous lines of tissue/cell-specific Aldh2 KO mice by crossing Aldh2^f/f mice with corresponding Cre-expressing mouse lines (Fig. 1a; Extended Data Fig. 1a). The Aldh2^f/f line was further validated by creating an E2a-Cre germline knockout which showed comparable serum AcH increases compared to global Aldh2 KO mice post ethanol administration (Extended Data Fig. 1b). The deletion of ALDH2 in tissue/cell specific Aldh2 KO mice was confirmed by western blot or immunostaining analyses (Extended Data Fig. 1c). Next, we measured serum AcH in various organ-specific Aldh2 KO mice post ethanol gavage. As illustrated in Fig. 1b, global deletion of the Aldh2 gene (Aldh2^E2a−/−) elevated serum AcH by 18.11-fold, while deletion of the Aldh2 gene in hepatocytes (Aldh2^Hep−/−) and adipocytes (Aldh2^Adipo−/−) resulted in serum AcH elevation by 2.88-fold and 1.83-fold, respectively. There is a tendency of serum AcH elevation in intestinal epithelial cell-specific Aldh2 KO mice (Aldh2^{Villin−/−}) compared to WT mice, but it did not reach statistical significance. Surprisingly, deletion of the Aldh2 gene in endothelial cells (Aldh2^Tek−/−), smooth muscle (Aldh2^{MyhErt−/−}), skeletal muscle (Aldh2^Mck−/−), skin (Aldh2^{K14Ert−/−}), bile duct cells (Aldh2^{Sox9Ert−/−}), kidney (Aldh2^{Podocin−/−}), and lung (Aldh2^{SftpcErt−/−}) did not elevate serum EtOH or AcH levels, which is consistent with no differences in drinking preference between these KO mice and their corresponding controls (Extended Data Fig. 2, Supplementary Fig. 1).

Fig. 1: — a, Generation of 10 lines of tissue/cell-specific *Aldh2* knockout (KO) and E2a-Cre germline *Aldh2* KO mice (Created with Biorender.com). b,Comparison of serum acetaldehyde (AcH) and ethanol (EtOH) levels of germline and specific single organ *Aldh2* KO mice (Cre⁺ groups) with their littermate control (Flox groups) mice 3 hours after ethanol (5g/Kg) gavage (n=5 in each group; striped square stands for the missing sample). Fold change and p values were shown on the right. (**p=0.0079, **p=0.0079,*p=0.0317, *p=0.0159) c, Serum EtOH and AcH levels in male and female WT (*Aldh2* ^f/f) mice (male: n=10, n=8; female: n=7, n=9), *Aldh2* ^Hep−/− (male: n=9, n=9; female: n=8, n=8), *Aldh2* ^{villin−/−} (male: n=9, n=7; female: n=9, n=7), and *Aldh2* ^{Hep−/−Villin−/−} mice (male: n=7, n=8; female: n=8, n=7) 1 hour (dose: 2g/Kg) and 3 hours (dose: 5g/Kg) post ethanol gavage (upper left *p=0.0172,**p=0.0054,***p=0.0007; upper right *p=0.0477, ***p=0.0009, ***p=0.0002, *p=0.0468; lower left *p=0.0439, *p=0.0117,**p=0.0069; lower right: *p=0.0260, ***p=0.0009, **p=0.0011, *p=0.0117). Values represent means±SEM. *p<0.05, **p<0.01, ***p<0.001. Two-sided Student’s t-test and one-way ANOVA were used for the comparison between two groups. ns: No significance.

The fact that deletion of Aldh2 in individual organs led to only slight or no elevation in serum AcH compared to ∼20-fold elevation in global Aldh2 KO mice may suggest the redundance of ALDH2 expression in multiple organs for systemic AcH clearance. To further explore this hypothesis, we generated liver and gut Aldh2 double KO (Aldh2^{Hep−/−Villin−/−}) (dKO) mice since both the liver and gut express high levels of ALDH2 (Extended Data Fig. 3a). The ALDH2 protein deletion in the Aldh2 dKO mice was confirmed by immunostaining analysis (Extended Data Fig. 3b). These dKO mice were treated with ethanol gavage, serum AcH levels were subsequently measured. The data in Fig. 1c revealed that double deletion of Aldh2 in the liver and gut resulted in much higher elevation of serum AcH than single deletion of Aldh2 in the liver or gut post oral administration of low (2g/kg) or high (5g/Kg) doses of EtOH in both male and female mice, suggesting that gut and liver ALDH2 synergistically promote systemic AcH clearance.

AcH is considered very reactive and toxic, and is generally believed to promote alcohol-associated liver disease²⁰, but our previous studies showed that liver injury (especially serum ALT levels) was not greater and even was lower in global or liver-specific Aldh2 KO mice after acute or chronic ethanol intake^18,21. Here we also compared various liver injury markers in WT and Aldh2 dKO mice post chronic-plus-binge ethanol feeding. Interestingly, compared to WT mice, Aldh2 dKO mice presented with slightly but not significantly decreased injury (serum ALT) and less infiltration of inflammatory cells in the liver (Extended Data Fig. 3c-d; Extended Data Fig. 4), which is consistent with our previous findings^18,21. Liver expression of genes involved in injury, inflammation, and fibrosis, gut epithelial tight junction-related genes, and gut permeability measured with FITC-dextran were decreased or unchanged in the gut and/or liver Aldh2 KO mice compared to WT controls after ethanol feeding (Supplementary Fig. 2), providing further evidence that these mice do not exhibit increased liver injury. The mechanisms by which Aldh2 dKO mice do not develop increased liver injury after chronic-plus-binge or acute ethanol binge challenge compared to WT mice are not clear. It is reasonable to speculate that elevated AcH in Aldh2 dKO mice activates compensatory protective mechanisms that ameliorate liver injury.

Bile efflux mediates AcH secretion from liver into gut lumen

Hepatocytes are highly polarized with sinusoidal (basolateral) and canalicular (apical) membranes that are separated by tight junctions. Hepatocytes not only secrete contents across the sinusoidal membrane into circulation but also excrete contents across the canalicular membrane into the bile duct, which flows into the gallbladder and is then released into the duodenum^22–24. Therefore, we wondered whether liver-derived AcH is also secreted into the gut via the bile efflux. To test this hypothesis, we measured bile and serum AcH in ethanol-binged mice. Intriguingly, we found that bile AcH levels were ∼5-fold higher than serum AcH levels, while ethanol levels were comparable in bile and serum post ethanol gavage (Fig. 2a). A similar result of much higher bile AcH levels than serum AcH levels was also obtained after intraperitoneal (IP) injection of ethanol (Extended Data Fig. 5a). Furthermore, we observed the enlarged gallbladder volume after ethanol gavage (Fig. 2b; Extended Data Fig. 5b). Then we measured local gut AcH levels from different parts of gut and found that significant amounts of AcH were detected in gastrointestinal (GI) lumen with a gradual decrease from the highest level in the duodenum to the lowest level in colon post ethanol gavage (Fig. 2c).

To explore why the duodenal lumen had the highest levels of AcH among the small intestine as shown in Fig. 2c, we speculated that AcH generated in the liver is secreted into bile that is subsequently drained into duodenal lumen. To test this hypothesis, we performed a bile duct ligation (BDL) experiment that blocks bile secretion from the liver to duodenal lumen. To rule out the detrimental effect of long-term BDL on the liver, we treated the mice with ethanol gavage shortly (3 hours) after BDL. Our data indeed revealed that BDL markedly reduced AcH levels in duodenal lumen (Fig. 2d). Additionally, we found that liver Aldh2 KO (Aldh2^Hep−/−) mice have higher AcH levels in bile and duodenal luminal content than WT mice, which is consistent with higher liver AcH levels measured in Aldh2^Hep-/ mice vs. WT mice post ethanol gavage (Fig. 2e).

AcH is extremely reactive, forming both reversible and irreversible protein adducts and chemical condensates^25–27. For example, one AcH metabolite, 2-methyl-thiazolidine-4-carbonyl-glycine (MTCG), is formed via the breakdown of glutathione (GSH) to cysteinylglycine (CysGly) which reacts with AcH and has been found in bile after ethanol administration to rats (Extended Data Fig. 5c)^28,29. To determine whether the AcH measured in bile was due to this metabolite, we depleted glutathione using buthionine sulfoximine (BSO) and treated the mice with ethanol gavage (Extended Data Fig. 5d,g) ³⁰. As illustrated in Extended Data Fig. 5e-i, two different BSO treatment protocols did not affect bile AcH levels, suggesting that the AcH in bile is unlikely to be in the form of MTCG. BSO did reduce serum AcH levels in one experimental paradigm but not the other. A previous study also reported that treatment with BSO did not affect serum AcH in rats³¹. The reason for the inconsistent effect of BSO treatment on serum AcH levels is not clear, which may be due to the different BSO treatment timing and different EtOH dosing. In summary, our results showed that BSO treatment did not affect bile AcH levels, suggesting that acute reductions in glutathione do not influence bile AcH levels, and the AcH in bile is not in the form of MTCG.

Gut AcH clearance is not affected by gut microbiota

To determine how gut AcH is cleared, we first explored whether AcH in intestinal lumen is metabolized by gut microbiota given the implications of microbiota in ethanol metabolism, although these previous findings are inconsistent^32,33. To define the role of gut microbiota in AcH clearance, we used germ-free (GF) mice that are not naturally colonized by microorganisms in gut. Our data revealed that except for duodenum, AcH levels in other intestinal segments, serum, bile, and liver tissue, were comparable between GF and specific pathogen-free (SPF) control mice post ethanol administration (Fig. 3a). EtOH levels were similar between GF and SPF mice (Fig. 3a). Additionally, liver ALDH2 protein expression levels were similar between GF and SPF control mice (Extended Data Fig. 6a). Interestingly, we noticed that GF mice produce more bile than control mice (Extended Data Fig. 6b), which could contribute to the higher levels of duodenal lumen AcH in GF mice post ethanol administration. To further prove this possibility, we treated C57BL/6N mice with antibiotics cocktail to thoroughly eradicate the effects of gut microbiota, and we found such short-term depletion of gut bacteria did not significantly affect bile production volume. Then the antibiotics-treated mice were gavaged with ethanol, and EtOH and AcH levels were measured in bile, serum, and intestines. No differences were observed between antibiotics-treated mice and control mice (Fig. 3b).

Fig. 3: — a, AcH and EtOH levels in gastrointestinal (GI) lumen, bile, liver, and serum from male germ-free (GF) (n=7) and specific pathogen free (SPF) (Control) (n=7) mice 3h post ethanol gavage (5g/kg) (*p=0.0306). b, C57BL/6N mice were treated with antibiotics cocktail to thoroughly eradicate the effects of gut microbiota (Diagram created with Biorender.com). EtOH and AcH levels in serum, bile and intestinal luminal content in control (n=6, n=5, n=5) and antibiotics-treated mice (n=5, n=5, n=5) 3h post ethanol gavage (5g/kg). c, A diagram for intra-intestinal AcH injection (20mM) in SPF (Control) and GF mice (Created with Biorender.com), and luminal and portal blood AcH levels were measured by GC-MS (n=6).

The luminal content AcH levels presented in this figure represents the diluted concentrations (see details in methods). Values represent means±SEM. *p<0.05, Two-sided Student’s t-test was used for the comparison between indicated two groups.

Next, to elucidate the direct effect of gut microbiota in luminal AcH metabolism, we fasted GF mice and control mice for 12 hours, and subsequently performed intra-intestinal injection of AcH, followed by luminal content and portal blood collection for AcH measurement, and we found no differences in AcH levels between GF and control mice (Fig. 3c).

Gut AcH is cleared via gut/liver ALDH2 first-pass metabolism

To determine whether gut ALDH2 is involved in luminal AcH clearance, we used gut epithelial Aldh2 KO (Aldh2^{Villin−/−}) mice in the following experiments. First, we found that compared to WT mice, Aldh2^{Villin−/−} mice had comparable ethanol levels but much higher AcH levels in duodenal lumen, portal blood, and liver tissues post ethanol gavage (Fig. 4a), which implied the key role of gut ALDH2 in AcH metabolism. Second, to explore the fate of AcH in gut, we performed intra-intestinal AcH injection and subsequently measured gut luminal and portal AcH levels. As illustrated in Fig. 4b, c, after intra-intestinal AcH injection, Aldh2^{villin−/−} mice had higher portal blood and duodenal luminal AcH levels than WT mice; meanwhile, portal blood AcH levels were detected ~10-fold lower than those in duodenal luminal content (Fig. 4c), suggesting that luminal AcH is mainly metabolized by gut epithelial ALDH2, but can be absorbed through portal blood with low efficiency. Finally, we conducted intra-intestinal AcH injection in liver Aldh2 KO (Aldh2^Hep−/−) mice. Interestingly, liver and bile AcH levels were higher in Aldh2^Hep−/− mice than WT mice post intra-intestinal AcH administration (Fig. 4d), demonstrating that gut luminal AcH is reabsorbed back to the liver to be further metabolized by liver ALDH2.

Fig. 4: — a, EtOH and AcH levels in duodenal luminal content, portal blood, and liver tissue from gut epithelium specific *Aldh2* KO mice (*Aldh2* ^{Villin−/−}) (n=6) and control mice (*Aldh2* ^f/f) (n=4) 3h post ethanol gavage (5g/Kg) (**p=0.0047,***p=0.0009,*p=0.0311). b, A diagram of *in vivo* intra-intestinal AcH solution (20mM) injection in C57BL/6N mice (Created with Biorender.com). c, Portal blood (n=4, n=5) and duodenal luminal (n=6) AcH levels of *Aldh2* ^{Villin−/−} and control mice (*Aldh2* ^f/f) mice 5 min and 15 min post intra-intestinal AcH (20mM) injection (*p=0.0132, **p=0.0029, *p=0.0271, **p=0.0018). d, AcH levels in liver tissue (n=6, n=5) and bile (n=4, n=4) from liver specific *Aldh2* KO mice (*Aldh2* ^Hep−/−) and control mice (*Aldh2* ^f/f) 15 min post intra-intestinal AcH (20mM) injection (*p=0.0131, *p=0.0361).

The luminal content AcH levels presented in this figure represents the diluted concentrations (see details in methods). All male mice were used. Values represent means±SEM. *p<0.05, **p<0.01, ***p<0.001. Two-sided Student’s t-test was used for the comparison between indicated two groups. ns: No significance.

Bile flow dynamics regulate blood and liver AcH clearance

Hepatocytes can secrete contents into bile and circulation via their apical and basolateral membranes, respectively^22,23. To define how much liver-generated AcH is secreted into bile and circulation, we performed ex vivo liver perfusion with ethanol, collected the perfusate from the portal vein and bile from gallbladder, and AcH levels were measured. As illustrated in Fig. 5a, the relative AcH levels in bile and the perfusate were 35.95% and 64.05%, respectively, suggesting that ∼30% and ∼70% of secreted AcH from the hepatocytes enter into bile and circulation, respectively.

Fig. 5: — a, A diagram of liver perfusion with ethanol (25% vol/vol) in male C57BL/6N mice (left panel), and the total amount of AcH in liver perfusate collected from portal vein and bile collected from common bile duct were determined by GC-MS measurement (right panel) (n=5), and the percentage was calculated and is shown. b, AcH levels in liver tissue, serum, and bile from sham male mice (n=5) and mice received bile duct ligation (BDL) (n=5) were measured 3h post ethanol gavage (5g/Kg) by GC-MS (***p=0.0008, **p=0.0041, **p=0.0020). **c,d,** AcH levels in liver tissue, serum, bile samples from male *Bsep* KO mice (*Bsep*^−/−) (n=7, n=6), *Mdr2* KO mice (*Mdr2*^−/−) (n=5, n=6), and their littermate control mice (shown as ‘Wild type (WT)’) were measured 3h post ethanol gavage (5g/Kg) (c:*p=0.0168, *p=0.0174; d:*p=0.0303,*p=0.0348,*p=0.0182). e, AcH levels in liver tissue, serum, and bile samples at different time points (1h, 3h, and 6h) from male C57BL/6N mice fed with chow diet (n=5) and ursodeoxycholic acid (UDCA) diet (n=5) were measured 3h post ethanol gavage (5g/Kg) (*p=0.0187, *p=0.0134, *p=0.0129). f, A diagram of bile flow regulation by different inhibitors (Created with Biorender.com). **g-j,** AcH levels in liver tissue, serum, and bile samples from male C57BL/6N mice pre-treated with vehicle, or Cyclo-1 (n=3, n=4), Novobiocin (n=7, n=4), Quinidine (n=5, n=5) and Rifampicin (n=5, n=5), were measured 3h post ethanol gavage (5g/Kg) (g:*p=0.0411,**p=0.0048; h:*p=0.0377, *p=0.0350, **p=0.0029; i:*p=0.0132; j:**p=0.0020, *p=0.0167).

Values represent means±SEM. *p<0.05, **p<0.01. Two-sided Student’s t-test was used for the comparison between indicated two groups. ns: No significance.

Abbreviation in panel f: MRP: multidrug resistance-associated protein; ABCG2: adenosine triphosphate (ATP)-binding cassette efflux transporter G2; ABCB11: ATP binding cassette subfamily B member 11; MDR: multidrug resistance protein; BCRP: breast cancer resistance protein; BSEP: bile salt export pump.

To test whether bile flow affects AcH clearance, we examined the amount and distribution of AcH by modulating bile secretion and transportation. First, we found that mice received BDL had ∼2-fold higher levels of AcH in the serum and ~30% higher liver AcH levels compared to sham-treated mice (Fig. 5b). Second, we used multidrug resistance protein 2 KO (Mdr2 ^−/−) and bile salt export pump KO (Bsep^−/−) mice, two mouse models that are characterized by intrahepatic cholestasis and have reduced bile flow^34,35. As illustrated in Fig. 5c, d, compared to corresponding littermate control mice, Bsep^−/− mice had lower bile AcH but higher serum AcH levels, which was more noticeably observed in Mrd2^−/− mice. Third, we fed C57BL/6N mice with ursodeoxycholic acid (UDCA) diet for 1 month, which is known to elevate synthesis and secretion of bile³⁶. Our data revealed that compared to control diet fed mice, UDCA fed mice presented with lower liver and bile AcH levels (Fig. 5e). Furthermore, we treated C57BL/6N mice with several inhibitors targeting specific bile acid transporters (Fig. 5f), which are known to affect intrahepatic bile flow^37–40 and found that inhibition of bile flow or efflux elevated serum AcH levels (Fig. 5g-j). In contrast, serum, liver, and bile EtOH concentrations were not affected by the various interventions or treatments described above, such BDL, Mdr2 KO, Bsep KO, UDCA treatment, and several bile inhibitors besides rifampicin, which reduced serum EtOH concentrations (Extended Data Fig. 7). Finally, hepatic ALDH2 expression levels and enzymatic activities were comparable between BDL, Mdr2^−/−, Bsep^−/−, UDCA treated mice, and their corresponding control mice (Extended Data Fig. 8a-c), suggesting that altered hepatic and serum AcH in mice with these treatments are not due to alterations of liver ALDH2 activity and expression.

Bile flow dynamics control drinking behavior

To elucidate the significance of bile-excreted AcH in regulating drinking behavior, we evaluated ethanol consumption and preference in mice after modulating bile flow. First, UDCA diet or control diet-fed mice were subjected to drinking-in-the-dark (DID) paradigm to evaluate their binge-like drinking behavior (Fig. 6a)⁴¹. UDCA-fed mice consumed more alcohol in DID experiments than control diet-fed mice (Fig. 6b; Extended Data Fig. 8d). Then, we conducted 2-bottle choice (2-BC) assay in which the mice were simultaneously offered 2 bottles, one bottle contained drinking water while the other one contained escalating concentrations of alcohol (3%−15%). Our data revealed that UDCA-fed mice showed higher alcohol preference than control diet-fed mice (Fig. 6c). We also performed DID or 2-BC experiments in Mdr2^−/− mice and their control mice. Compared to littermate control mice, Mdr2^−/− mice presented with lower ethanol consumption in DID experiments, and lower alcohol preference at high ethanol concentrations (9%, 12% and 15%) in 2-BC assay (Fig. 6d, e; Extended Data Fig. 8d). Finally, we also did a DID experiment in Bsep^−/− mice and found that Bsep^−/− mice consumed lower amounts of alcohol than WT mice (Extended Data Fig. 8e).

Fig. 6: — a, A diagram of drinking in the dark (DID) assay and two-bottle choice (2-BC) assay. D1-D4: day 1 to day 4 (Created with Biorender.com). b, DID assay in male C57BL/6N mice fed with chow diet (n=10) and UDCA diet (n=10) (*p=0.0332, *p=0.0190, *p=0.0370). c, 2-BC assay in male C57BL/6N mice fed with chow diet (n=10) and UDCA diet (n=10). Upper panel: alcohol intake curves and statistics for relative area under curve (AUC). Lower panel: alcohol preference (%) curve and statistics for relative AUC (***p<0.0001, ***p=0.0002, **p=0.0021, ***p=0.0002, * p=0.0117, ***p=0.0004, *p=0.0156, *p=0.0443, *p=0.0409). d, DID assay in male *Mdr2* KO mice (*Mdr2*^−/−) (n=10), and their littermate control mice (WT) (n=11) (*p=0.0205, **p=0.0036, *p=0.0383). e, 2-BC assay in male *Mdr2*^−/− and WT mice (n=10). Upper panel: alcohol intake curves and its relative AUC. Lower panel: alcohol preference (%) curve and its relative AUC (upper: *p=0.0236, **p=0.0010, **p=0.0004, *p=0.0136, *p=0.0408, *p=0.0230, **p=0.0052, **p=0.0049, *p=0.0126).

Values represent means±SEM. *p<0.05, **p<0.01, ***p<0.001. Two-way ANOVA and two-sided Student’s t-test were used for the comparison between indicated two groups. ns: No significance.

Liver and gut ALDH2 synergistically promote alcohol drinking

The above data clearly indicated that AcH in liver is efficiently removed through bile efflux and is further metabolized by ALDH2, while a small proportion of AcH in gut luminal content is reabsorbed via portal vein back to liver, suggesting that the liver and gut synergistically clear AcH. To further confirm this notion, we measured portal blood, bile, and duodenal luminal AcH levels in Aldh2 dKO or single KO mice. As illustrated in Fig. 7a, compared to wild-type control mice, liver Aldh2 KO mice had higher levels of bile and duodenal AcH levels, but comparable portal AcH levels; while gut Aldh2 KO mice had higher levels of portal blood, bile and duodenal AcH levels. The dKO mice had the highest levels of AcH in portal blood, bile, and duodenal lumen among these four groups (Fig. 7a; Extended Data Fig. 9a). In contrast, EtOH levels in cerebellum, liver, duodenal lumen, and bile samples were similar among the four groups (Extended Data Fig. 9b).

Next, we also measured cerebellar AcH levels. As illustrated in Fig. 7a, compared to WT (Aldh2^f/f) mice, Aldh2^Hep−/− and Aldh2 dKO mice had higher cerebellar AcH levels post administration of 5g/kg ethanol with higher levels in dKO than in Aldh2^Hep−/− mice. Moreover, we also observed higher AcH levels in cerebellum, liver, duodenal lumen, and bile samples in dKO mice than those in WT mice post administration of a lower dose of ethanol (2g/kg) (Extended Data Fig. 9c). Interestingly, cerebellar AcH levels had positive correlation with serum AcH levels (Extended Data Fig. 9d). Meanwhile, AcH in other brain regions were also measured, including prefrontal cortex (PFC), hippocampus, thalamus, and hypothalamus. The data revealed that PFC and hippocampus also exhibited significant increases in AcH, while the thalamus and hypothalamus did not have elevated AcH (Extended Data Fig. 9e, f).

To test whether the different levels of systemic and central AcH affect metabolic rates and drinking behavior among the four groups, we performed metabolic chamber tests, DID and 2-BC experiments. Our data revealed that compared with WT mice, dKO mice had remarkable suppression of respiratory quotient, carbohydrate oxidation (Fig. 7b), physical activity, and food or water intake (Extended Data Fig. 10). Liver Aldh2 KO mice had a trend of reduced metabolic rates compared to WT mice, however, the difference did not reach statistical significance, while gut Aldh2 KO mice had similar metabolic rates as WT mice (Fig. 7b; Extended Data Fig. 10).

Elevation of peripheral AcH is believed to prevent further alcohol drinking due to its unpleasant toxic effect⁴². In agreement with this notion, our data revealed that intraperitoneal injection of AcH led to reduced alcohol preference (Supplementary Fig. 3a-b). To further explore whether dKO mice exhibit altered alcohol preference relative to Aldh2^f/f, Aldh2^Hep−/− or Aldh2^{villin−/−}, these four groups of mice were subjected to a 2-BC assay. We found alcohol consumption was comparable in Aldh2^{villin−/−} and Aldh2^f/f mice. Aldh2^Hep−/− mice showed lower alcohol intake than Aldh2^f/f control at 12% and 15%, and dKO mice consumed much less alcohol compared to other groups (Fig. 7c, d). The significant reduction of alcohol preference in dKO is specific to alcohol because no difference was found in sucrose solution (2% and 5%) consumption among these four groups of mice (Supplementary Fig. 3c). In addition, WT and dKO mice exhibited similar latency-to-fall after EtOH gavage in the rotarod test, indicating that the dKO mice exhibit similar sensitivity to the locomotor effects of EtOH (Supplementary Fig. 3d). Finally, we also performed DID assay and found that male dKO mice had the lowest ethanol intake among four groups of male WT, Aldh2^Hep−/−, Aldh2^{villin−/−}, and dKO mice (Fig. 7e). Interestingly, female dKO mice also had the lowest alcohol preference and intake among the four groups of female mice (Supplementary Fig. 4).

Discussion:

In the current study, we identified a liver-gut ALDH2 loop that plays an important role in promoting systemic AcH clearance and controlling drinking behavior (Fig. 8). ALDH2 has been investigated for decades as a therapeutic target for the treatment of AUD^43,44. Disulfiram (Antabuse), a potent ALDH2 inhibitor, is an FDA approved drug for AUD treatment but its use has been limited due to its side effects, patients’ compliance, and its nonspecific inhibition of other ALDHs⁹. Thus, tissue specific ALDH2 inhibition and selective ALDH2 inhibitors have been recently tested in pre-clinical models¹⁸ and in patients^10,45. The redundance of ALDH2 expression in multiple organs for AcH metabolism as demonstrated in the current study suggests that targeting ALDH2 in an individual organ for AUD may not be efficient. Indeed, genetic deletion of the Aldh2 gene in the liver only prevents heavy drinking without affecting moderate drinking¹⁸, while genetic deletion of the Aldh2 genes in several other individual organs tested in the current study did not significantly affect alcohol drinking preference in mice. ALDH2 siRNA to knockdown liver ALDH2 was tested in human subjects and had very mild effects on blood AcH levels and AcH associated clinical symptoms, including flushing, nausea, and impacts on heart rate and blood pressure⁴⁵. Our current study revealed that genetic deletion of both the liver and gut Aldh2 gene synergistically attenuated drinking preference and alcohol consumption, suggesting targeting both liver and gut ALDH2 likely generates better therapeutic outcomes than targeting liver or gut ALDH2 alone, and that both gut and liver ALDH2 inhibition is necessary for ALDH2-targeted AUD therapeutics.

Fig. 8: — Alcohol is metabolized via ADH into AcH in hepatocytes. Approximately 70% secreted AcH from hepatocytes enters into circulation, while 30% secreted AcH is drained via the bile flow into duodenal lumen, where AcH is detoxified and cleared by gut epithelium ALDH2, while a small portion of AcH is resorbed back to liver for further metabolism by liver ALDH2. This ‘liver-gut ALDH2 loop’ promotes systemic AcH clearance, and more importantly, controls alcohol preference and drinking behavior. Targeting liver-gut ALDH2 loop may represent a novel therapeutic approach for AUD. (Created with Biorender.com).

Another important implication from the current study of peripheral Aldh2 deletion may help us better understand the effect of peripheral AcH on drinking behavior. There has been a long-running controversy over whether peripherally-produced AcH can pass the blood-brain barrier (BBB), thereby affecting drinking behavior⁴⁶. Early studies indicated that peripheral AcH was metabolized and cleared via ALDH at the BBB, preventing it from entering the brain^47–49. However, in the current study, we found that selective deletion of peripheral Aldh2 in the liver and gut elevated both serum and central AcH concentration in several brain regions and reduced drinking preference compared to wild-type mice, suggesting that peripherally-generated AcH can cross the BBB to modulate drinking behavior. Additionally, we have previously demonstrated that ALDH2 deficiency in forebrain neurons (Aldh2^{Camk2a−/−}) or glial cells (Aldh2^Gfap−/−) did not significantly alter the level of brain AcH and alcohol drinking behaviors^18,19, and astrocytic ALDH2 deletion regulated both ethanol- and acetate-induced cellular and behavioral effects on the brain by activating distinct GABA synthesis pathways without affecting drinking preference¹⁹. Peripheral Aldh2 deletion did not alter rotarod performance after ethanol gavage in a single cohort of mice, which is in contrast to the observed increases in ethanol sensitivity for mice and humans with the ALDH2*2 polymorphism^50–52. These differences in behavioral response between organ-specific and full body ALDH2 deficiencies should be explored in future studies. Collectively, these findings suggest that selective, peripheral ALDH2 inhibition may be sufficient to drive higher levels of AcH in the brain, making it much easier to develop therapeutics by targeting peripheral ALDH2. However, how elevated peripheral and/or central AcH via the inhibition of peripheral ALDH2 reduces drinking preference and behavior remains largely unclear. A previous study by Mews et al. suggests that the ethanol metabolite acetate plays an important role in controlling epigenetic processes in the brain, which is associated with alcohol-induced behavioral adaptations⁵³. Inhibition of ALDH2 is known to diminish acetate production, so it will be very interesting to see whether inhibition of peripheral ALDH2 will abolish alcohol metabolite acetate-mediated epigenetic changes in the brain and alcohol-induced behavioral adaptations observed by Mews et al.⁵³.

Enterohepatic circulation is a process through which hepatocytes secrete bile acids, toxins, and metabolites into the gut via the bile secretion, some of these compounds will re-absorb back to the liver via the portal circulation, playing an important role in detoxification⁵⁴. In the current study, we found that AcH was detected in the bile from the ethanol-treated mice with much higher levels than serum AcH concentration. It is difficult to calculate the exact percentage of liver AcH secretion into circulation and bile in vivo but our ex vivo perfusion experiments revealed that ∼70% of the secreted AcH from hepatocytes entered into circulation while ∼30% entered into bile. With bile excretion into the gallbladder being an important clearance mechanism for ethanol-derived AcH, the gallbladder may be at increased risk for AcH-mediated cancer development. One study found that increased ethanol intake was associated with a 60% increased risk for gallbladder cancer^55,56, which may be attributed to high AcH levels in gallbladder post alcohol intake as demonstrated in the current study. Another interesting clinical observation was that alcohol intake is associated with a decreased risk of developing symptomatic gallstone disease^57–59. However, it could be the opposite that is true: gallstone disease may reduce ethanol consumption⁶⁰ because of impaired AcH clearance via the bile. Therefore, the unexpected finding of high bile AcH levels as demonstrated in the current study may partially explain alcohol-associated biliary tract injury and disease, which deserves further study.

We also found that among the different segments of intestine, duodenal lumen had the highest levels of AcH, which were partially drained from the gallbladder. Furthermore, our data suggest that AcH in small intestine is further metabolized by gut ALDH2 as gut luminal AcH levels were much higher in gut-specific Aldh2 KO mice than those in WT mice post ethanol gavage or intra-intestinal AcH injection. In contrast, gut microbiota likely plays a minor role in gut AcH clearance as depletion of gut microbiota (germ-free or antibiotic treatment) did not affect gut AcH levels. Additionally, we also tried to determine whether and how much gut luminal AcH can be re-absorbed back to the portal vein and circulation and whether these re-absorbed AcH is metabolized by liver ALDH2. We detected AcH in portal blood post intra-intestinal injection of AcH but its levels in portal blood were ~10-fold lower than those in intestinal lumen. Intriguingly, AcH levels were higher in the liver of Aldh2^Hep−/− mice than in WT mice post intra-intestinal AcH injection. Collectively, our data suggest luminal AcH can be absorbed back into the liver via the portal vein with low efficiency where it is further metabolized by liver ALDH2.

Finally, the important role of the enterohepatic circulation in systemic AcH clearance is further supported by several experiments showing that altering bile flow affected AcH secretion into bile or serum by inhibiting several bile acid transporters (via the genetic deletion of Mdr2 or Bsep2 genes, or chemical inhibitors) or by performing bile duct ligation. In addition, modulation of bile flow led to alterations in drinking behavior. Increasing bile flow via UDCA diet increased ethanol consumption while Mdr2 KO and Bsep KO mice exhibited decreased ethanol consumption. While the alteration in serum AcH in these models likely contributes to the observed alterations in drinking behavior, the potential regulation of drinking behavior by bile acids themselves cannot be discounted. Previous studies have shown that Mdr2 KO and Bsep KO mice exhibit opposite effects on bile acid profiles in the serum, liver, bile, and feces, so much so that knocking out Bsep in Mdr2 KO mice reverses liver injury⁶¹. Therefore, it would be unlikely that we would see similar decreases in ethanol consumption in both Mdr2 KO and Bsep KO mice if changes in bile acid composition were playing a dominant role. While changes in bile acid composition may not drive the changes in drinking behavior, changes in bile flow and overall bile acid signaling may contribute. Bile acids are implicated in a range of physiological processes including the control of food intake. Previous studies have demonstrated that bile acids promote food satiety and protect against obesity through multiple mechanisms, including Takeda G protein-coupled receptor 5 (TGR5, also known as G protein-coupled bile acid receptor 1 (GPBAR1) mediated activation of peripheral vagal afferent neurons⁶², alteration of energy balance via hypothalamic TGR5 activation⁶³, and TGR5-mediated secretion of appetite regulating peptides⁶⁴. Future studies should investigate whether these bile-induced satiety mechanisms also impact ethanol consumption.

Our findings suggest that enterohepatic clearance of AcH is an important mechanism of AcH detoxification that regulates ethanol consumption; however, ethanol metabolism differs between mice and humans¹. Mice metabolize ethanol about 5-fold faster than humans and generate AcH more quickly¹. How this basal difference in ethanol metabolism affects the aversive properties and clearance of systemic AcH in different species is not well understood. Therefore, this paradigm of AcH clearance should be further validated in future non-human primate and clinical studies.

Targeting ALDH-mediated AcH clearance has been a therapeutic strategy for AUD since the approval of disulfiram. However, the efficacy and long-term safety of this approach needs further study. First, AcH is classified as a group 2B carcinogen, and people with the ALDH2*2 polymorphism have higher risks of several different types of cancer⁶⁵. Second, individuals with the inactive ALDH2*2 allele generally have decreased ethanol consumption, but people with the ALDH2*2 allele comprise up to 20% of excessive drinkers in certain studies, indicating that the increases in AcH itself may only decrease ethanol consumption in a subset of the population^66–70. Thus, further studies are needed to clarify whether targeting ALDH2 is an effective and safe strategy for the treatment of AUD. Our data demonstrates that systemic AcH levels are a powerful regulator of drinking behavior in ethanol-naïve and non-dependent mice. Future studies utilizing more clinically-relevant preclinical models of AUD, such as aversion-resistant consumption and reinstatement after withdrawal in ethanol dependence, are needed to determine whether increasing systemic AcH will be efficacious in different subpopulations of AUD patients^71–73. Regardless, if short-term inhibition of ALDH2 and higher AcH levels lead to decreased ethanol consumption in AUD patients, this may provide a net benefit since ethanol itself also causes significant harm. One second-generation ALDH2 inhibitor, ANS-6637 (CVT-10216), was able to decrease ethanol consumption and seeking in ethanol-preferring rats, potentially by reducing the rewarding properties of ethanol⁷⁴. This inhibitor was also tested in healthy volunteers who reported decreased liking of alcohol but was discontinued in Phase 2 trials due to liver toxicity (NCT03970109)⁷⁵. Future testing of specific ALDH2 inhibitors in preclinical models and the clinic may shed light on how AcH regulates drinking behavior in different populations. Additionally, comparing central and peripheral ALDH2 deletion, or comparing similar ALDH2 inhibitors that are centrally bioavailable or peripherally-restricted would provide a great tool to determine how peripheral vs. central clearance of AcH contributes to drinking behavior, which may help identify better ALDH2 targeting therapies for AUD.

Methods:

Animals:

All animal experiments were approved by the NIAAA animal care and use committee. Animals were housed under a 12-h light/dark cycle with room temperature 22ºC ± 1ºC and humidity of 30–70%. Mice had free access to food and water unless otherwise specified. Mice between 8 and 24 weeks old were used for all experiments. Sex was not considered as a special factor in the study design but key experiments were replicated in both male and female mice. No statistical methods were used to pre-determine sample sizes but our sample sizes are similar to those reported in previous publications^18,19. Aldh2^flox/flox mice were described previously.¹⁸ The following Cre lines were used to cross with Aldh2^flox/flox via several steps to generate tissue/cell-specific Aldh2 knockout mice (Supplementary Table 1). Littermate controls were used in all experiments with organ-specific Aldh2 knockout mice. Animals were assigned randomly to the groups in all experiments to prevent selection bias.

For ERT inducible Cre mice, tamoxifen (dissolved in corn oil at 20mg/ml) (Sigma) was administered at the dose of 50mg/kg by i.p injection in a total volume not to exceed 0.1 mL to both male and female mice, once a day for not more than 5 days.

Other knockout mice used in the current study include: Mdr2^−/− mice (JAX 002539), Bsep ^−/− mice (JAX 004125, kindly provided by Victor Ling - British Columbia Cancer Agency)⁷⁶. Heterozygous mice were used for breeding to generate knockout and WT littermate controls.

Germ free mice were obtained from the NIAID in-house breeding colony, and SPF C57BL/6N mice (controls) were purchased from Taconic.

Acute ethanol gavage:

Mice received an oral gavage of up to 5 g/kg ethanol with 20G gavage needle (Catalog #: AFN2038C). Mice were placed on a heating pad at 38°C throughout the experiment to prevent hypothermia, and then serum or any other tissues were collected.

Acute ethanol intraperitoneal injection:

The appropriate amount of 4 g/kg ethanol was injected into the lower left area of mouse abdomen with 25G needle 1mL syringe (Monoject™).

Treatment of antibiotics:

Male C57BL/6N or C57BL/6J mice were treated, via oral gavage, with a cocktail of antibiotics (Ampicillin + Neomycin + Metronidazole + Vancomycin + Gentamicin1.0 + 1.0 + 1.0 + 0.5 + 1.0 g/mL) (200 μl/day) daily for 3 days to remove the host gut microbiota. This method was used previously by Kelly et al 2015, showing nearly complete reduction of gut bacteria ⁷⁷. At the day 4, mice were orally binged with ethanol (5g/Kg) and various tissues (including serum, liver, bile, portal blood, and intestinal luminal contents) were collected 1hour and 3hours post ethanol administration.

UDCA treatment:

Mice were fed with either a control diet consisting of standard chow (DYET#611559), or the same diet containing UDCA (0.5% wt/wt) for 3 weeks and body weight was measured prior to initiation of drinking experiments. Mice were subjected to DID or 2-BC experiments while the UDCA/control diets continued. At the end, mice were maintained on the UDCA diet for two weeks, and then were subjected to oral ethanol gavage and blood ethanol and AcH levels were measured. The 0.5% UDCA (Sigma-Aldrich, St. Louis, MO) supplemented chow compounded were purchased from Dyets animal diets (DYET#615093).

Bile/serum sample collection and preparation for EtOH/AcH measurement:

Bile was taken with 0.5 mL insulin syringe with 30G needle (BD 328466–1) at different timepoints from the gallbladder after ethanol gavage. The bile samples were collected with 1.5mL Eppendorf tube on ice, which were then frozen in liquid nitrogen for long-term storage. For serial serum samples, a micro‐hematocrit capillary tube was used to collect the blood from the mouse eye (retroorbital sinus) within 10s, then the blood was collected with a serum collection tube (Microvette® 500 Z Gel) and placed on ice for 10min. Next, samples were centrifuged 4min at 10,000g, 4°C. Each serum sample was quickly split into1.5ml Eppendorf tubes with aliquots of 25ul/tube within 10s. Aliquots were kept on ice or quickly frozen in liquid nitrogen.

Before the EtOH/AcH measurement, the serum/bile samples were thawed on ice for 10min. Bile samples (5uL) were transferred to the 20ul ice cold PBS solution in a 1.5mL Eppendorf tube on ice for dilution. Then, EtOH /AcH measurement of all samples were performed with gas chromatograph-mass spectrometer (GC-MS) analysis ¹⁹.

Measurement of AcH/EtOH in intestinal luminal content:

Due to the technical difficulty of collecting intestinal luminal content for direct EtOH/AcH measurement, we conducted measurement in this way: small or large intestine tissues were promptly collected after ethanol gavage, 1cm length proximal segment of duodenum (adjacent to pylorus) and 3cm proximal segment of mid small intestine, terminal ileum, and colon were collected with 1.5mL Eppendorf tube, then the intestinal lumen was fully exposed, and rinsed with 0.5mL cold PBS to make sure all luminal content resolved. All procedure was promptly performed on ice. The 30ul aliquots of PBS solution containing the luminal content were then frozen in liquid nitrogen for long-term storage and EtOH/AcH measurement. EtOH /AcH measurement of all samples were performed with gas chromatograph-mass spectrometer (GC-MS) analysis ¹⁹.

Treatment with different bile transport inhibitors:

Mice were treated with a single dose of the following inhibitors, 30 min later, mice received oral ethanol gavage (up to 5 g/kg), and serum and organs were collected 3h post ethanol gavage. Cyclosporin A (Pgp, MRP-1, BCRP, and LRP inhibitor, Cat: 30024–25MG, Sigma-Aldrich)(dissolved in olive oil, 25 mg/kg, IP)⁷⁸. Quinidine (ABCB1/MDR1/P-gp inhibitor, Cat: PHR3233–1G, Quinidine sulfate, Sigma Chemical) (100 mg/kg, IP)⁴⁰. Novobiocrin (ABCG2/BCRP/Mxr inhibitor, Cat: 46531–250MG, Novobiocin sodium salt, Sigma Chemical) (120 mg/kg, IP)⁴⁰. Probencid (ABCC2/MRP2 Inhibitor, Cat: 1563003–200MG, Sigma Chemical) (20 mg mg/kg, IP)⁴⁰. Rifampicin (ABCB11/BSEP Inhibitor, Cat: R3501–1G Sigma Chemical) (200 mg mg/kg, oral)³⁸. Ketoconazole (ABCB11/BSEP Inhibitor, Cat: PHR1385–1G, Sigma Chemical) (200 mg mg/kg, oral) ⁷⁹.

Bile duct ligation (BDL):

Ten to fourteen-week-old male C57BL/6N mice were randomly assigned to sham group and BDL group and were anesthetized by inhalation of isoflurane during the surgical operation. The mice were placed on a 38°C heating plate, abdominal fur was detached with an electric shaver. The abdominal skin was sterilized with iodine antiseptic solution followed by 70% ethanol for three times, then made an upper abdominal incision of ~2 cm and exposed the liver and common bile duct. For BDL group, the liver was lifted by using a moisturized saline cotton tipped applicators (Dynarex, Item# 4301), and the surgery was performed under microscope with separating the bile duct from the portal vein and hepatic artery by using micro-serrations forceps (McKesson), placing a 4–0 silk suture (Catalog #: ETHVCP392H) around the common bile duct, and secured with two surgical knots. Then, double ligation of the common bile duct was conducted with dissection between the two ligatures. After the surgical operation, the abdominal layers were closed with 4–0 sutures. The mice were placed on the heating pad until they were fully awake and active. To avoid the detrimental effect of long-term BDL on the liver, the mice received oral ethanol gavage (5g/Kg) 3 hours after BDL.

In vivo intra-intestinal AcH solution injection:

10–12 weeks old male C57BL/6N mice were anesthetized by inhalation of isoflurane during the surgery. First, a skin incision (1–1.5cm) was made on the abdominal wall along the midline starting from the xiphoid cartilage to expose the stomach and intestine. Then, the pylorus and distal colon were ligated with 4–0 silk suture (Catalog #: ETHVCP392H) and mini clips were used for intestinal obstruction. Then, each mouse received intra-intestinal injection of 1.5 mL acetaldehyde solution (20mM, prepared with PBS) (Sigma-Aldrich). The intestinal luminal liquid, portal blood and other samples were collected 5–15 min post-surgery.

Mouse liver ethanol perfusion:

10–14 weeks old male C57BL/6N mice were anesthetized by inhalation of isoflurane during the liver perfusion. The mouse was placed in the supine position, and a 3cm wide incision starting from the xiphoid process was made along the middle line of mouse abdomen, then, liver and portal vessel were well exposed. The inferior vena cava, common bile duct, and hepatic artery were carefully isolated and ligated with 4–0 silk suture (Catalog: #ETHVCP392H), and positioned another 4–0 silk suture under the arch of the portal vein proximal to the liver. The second silk suture distal of the inferior mesenteric vein distal from the liver was placed. Once the sutures are in place, cannulate the portal vein with a 22G catheter (Jelco® IV Catheters). Tie the first suture past the catheter tapper. And secure the lower portion of the catheter with the second suture. When the catheter is secured, insert a 3 mL syringe long into the catheter to inject 1mL ethanol (25% vol/vol) into the liver. Three minutes later, bile was taken from the gallbladder and perfusion liquid was collected from the catheter for AcH measurement.

Western blot:

Tissue lysates were prepared as previously described ⁸⁰. Liver and intestine tissues used in this project were homogenized by using RIPA lysis buffer (1mg tissue/60μL) containing protease inhibitors (Santa Cruz, CA) following the manufacturer’s instruction. Samples were processed on ice, and then were centrifuged at 13,200g for 10minutes at 4°C in a microcentrifuge. The concentrations of protein extracts were determined by using BCA Protein Assay Kit (Thermo Fisher, Waltham, MA, USA), then were mixed with loading buffer, and loaded equal amounts of protein samples into the wells of 4–12% Bis-Tris protein gels (Bio-Rad, Hercules, CA, USA). Transferring the proteins to nitrocellulose membranes (Thermo Fisher, Waltham, MA, USA). Protein bands were visualized and analyzed by using SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher, Waltham, MA, USA). The results were determined with ImageJ software (National Institutes of Health, Bethesda, MD). The primary antibodies used for analysis include anti-ALDH2 (1:1000; abcam, Cat#.:ab194587), anti-β-Actin (1:10,000; Sigma-Aldrich, Cat#. A1978), and anti-GAPDH (1:1000; Cell Signaling Technology, (14C10), Cat#. #2118L). Secondary horseradish peroxidase-conjugated antibodies (1:5000; Cell Signaling Technology, Cat.# 7076 (mouse) and 7074 (rabbit)) were used for analysis.

Immunofluorescent staining:

Immunofluorescent staining was described previously⁸¹. Briefly, formalin-fixed mouse samples were processed (Leica TP1020), paraffin embedded tissue sections (liver and intestine) were incubated in 3% H₂O₂ for 20 mins at RT. All slides were blocked in blocking solution (3% goat serum in PBS) for 1 hour at RT and were incubated with anti-ALDH2 (1:500 dilution, abcam, Cat#.: ab194587), Pan-keratin (C11) (1:50; Cell Signaling Technology, Cat.# 4545), or CD31 (1:100; Cell Signaling Technology, Cat.# 77699S) overnight at 4°C. After PBS washing for 3 times, slides were incubated with the fluorescence conjugated antibodies (1:1,000; Goat Anti-Rabbit 549 or 488 (Vector Laboratories, Cat.#: DI-1549 or DI1488), or Horse Anti-Mouse 488 (Vector Laboratories, Cat.#: DI-2488) for 1 hour at RT. Nuclear staining was obtained by incubation with 1 mg/mL 4’, 6’-diamino-2-phenylindole (DAPI) for 5 min at room temperature. Fluorescent images were obtained using a LSM 710 confocal microscope (Zeiss) running Zen 3.2 (Carl Zeiss Microscopy, White Plains, NY, USA) or a Leica Aperio Versa slide scanner (Versa software v1.0.4.; Deer Park, IL, USA).

Liver ALDH2 enzymatic activity measurement:

Fresh mouse liver tissue lysates were prepared with a Dounce homogenizer in cold PBS. Then, we suspended the homogenate to 25 mg/mL in PBS, and diluted homogenate by adding 4 volumes of Extraction Buffer to a sample protein concentration of 5 mg/mL. Next, the Mitochondrial aldehyde dehydrogenase (ALDH2) activity assay kit (Abcam; ab115348) was used to determine mitochondrial ALDH2 in liver samples by following manufacturer’s instructions, and finally, a standard microplate reader was used to record the absorbance of samples at 450nm. Data analysis was also performed by following the instruction.

Mouse model of chronic ethanol diet feeding

For the chronic ethanol induced liver injury assessment, the NIAAA mouse model⁸² developed by our lab was used in the current study. Generally, ethanol-fed or pair-fed mice were initially fed with control Lieber-DeCarli diet (Bio-Serv, Lot#289767.00) ad libitum for 5 days to acclimatize them to liquid diet and tube feeding. Then, the mice of ethanol group (EtOH group) were fed with ethanol Lieber-DeCarli diet (Bio-Serv, Lot#265380.00) containing 5% (vol/vol) ethanol for 10d and control groups are pair-fed with the isocaloric control diet. On Day 11, ethanol-fed and pair-fed mice were conducted with oral gavage with a single dose of ethanol (5g/kg) or isocaloric maltose dextrin solution, respectively. The mice were euthanized 9 h later, and tissues were collected for further analysis. More details about this protocol can be found in our previously published work⁸².

RNA isolation and real-time quantitative PCR (RT-qPCR)

Total RNA was extracted from liver and small intestinal tissues by using TRIzol reagent (Invitrogen, Carlsbad, CA) following the manufacturer’s instruction. One μg total RNA was reverse transcribed into cDNA by using High-Capacity cDNA Reverse Transcription kit (Thermo Fisher Scientific). RT-qPCR was performed by SYBR Green Realtime PCR master mix (ABM, BlasTaq™2XqPCR, Cat#:G892). The mRNA levels were determined by QuantStudioTM 6 Real-Time PCR System (278861830; Thermo Fisher Scientific). The expression levels of target genes were normalized to 18S rRNA expression. Comparative Ct (2-ΔΔCt) method was performed to quantify the mRNA expression level. All primers used for RT-qPCR are listed in Supplementary Table 2.

Immunohistochemistry (IHC) staining

The paraffin-embedded slides for immunohistochemistry staining were applied for heat-induced epitope retrieval first. For MPO (Primary antibody: BIOCARE MEDICAL Lot:102819), IBA-1 (EMD Millipore Cat.#: MABN92) and OPN (CST; Rabbit mAb #88742) staining, citrate buffer (Thermo Fisher, USA) was used for antigen retrieval. Then, the slides were incubated in 3% H₂O₂, and followed by another 60 mins incubation in 1% BSA. Sections were incubated with primary antibodies overnight at 4°C. ImmPACT AEC kit (Vector Laboratories, Inc., Burlingame, CA) and DAB Peroxidase Substrate Kit (Vector Laboratories, Inc., Burlingame, CA) were used to visualize the staining according to the manufacturer’s instructions. The percentages of positive area of Sirius red, IBA-1, and OPN staining were determined with ImageJ software (National Institutes of Health, Bethesda, MD). The number of MPO⁺ cells in specific areas were counted for statistical analysis.

Behavioral studies:

(1). Dynamic measurement of metabolic parameters:

Indirect measurements of metabolic parameters were performed on an Oxymax Metabolic Cage System (Columbus Instruments, Columbus, OH) in a dedicated animal housing room to avoid noise and disturbance. The Oxymax System is a non-invasive set-up that directly records respiratory exchange ratio (O2 and CO2), as well as mouse physical activity measured by infrared beams and detectors. Littermate mice of each group with similar body weight (less than 5% difference) were individually housed 3 days for habituation. Then, the mice received a single dose of 5 g/kg ethanol gavage, and a variety of real-time metabolic parameters (including respiratory quotient, carbohydrate oxidation, food/water intake activity, fat oxidation, ambulatory movements) were automatically recorded by the system. Cages remained unopened during the measurements. During the whole process, the mice had access to water and food ad libitum.

(2). Drinking in the dark (DID) assay:

To avoid the disturbance on the behavioral test, the four-day DID procedure was performed in a quiet, dedicated housing room on a standard 12-hour light/dark cycle. Standard chow diet and water were provided ad libitum. The mice were individually housed for 1–2 weeks prior to daily assessment of DID, and the body weight of each mouse was recorded before the experiment. Three hours after lights were turned off in the animal housing room, the water bottle was replaced by an ethanol-containing bottle (20% vol/vol) for two hours (days 1–3) and for four hours at day 4, at which point the amount of ethanol consumed is weighted and recorded. The ethanol-containing bottles were replaced by normal drinking water after each period. Bottle replacement was performed while a red lamp was temporarily turned on to prevent disruption of the mice’ circadian rhythm.

(3). Two-bottle choice (2BC) assay:

Similar to DID assay, 10–14 weeks old mice were individually housed for 1–2 weeks prior to 2BC assay. After habituation, mice were given free choice between 2 same bottles, one of which containing normal drinking water while the other one containing escalating alcohol solutions prepared with drinking water (typically 3%, 6%, 9%, 12%, and 15% or 5%, 10%, 15%, and 20% vol/vol) or sucrose for sucrose preference experiments at 2% and 5% (weight/vol). The mice received each concentration for 4 days. Drinking volumes were measured daily, and bottle positions were interchanged to prevent learned preference ¹⁸.

(4). Rotarod Test

Rotarod test was performed as described previously with minor modifications ¹⁹. We utilized an accelerating rotarod paradigm (4 to 40 rotations per minute over 300s; ENV-575M, MED Associates) where each trial ended when mice fell off the rotarod or reached the full 300s measured by latency to fall. Mice underwent three training sessions per day for three days during the week prior to testing, and had to stay on the rotarod for at least 250s during testing to be considered eligible for testing. On testing days, mice were tested at baseline prior to ethanol administration, given either a 2 or 3 g/kg ethanol gavage, and rotarod fall latency was measured every 30 min for 90–120 minutes.

(5). Sucrose preference test

To evaluate animals’ interest in a sweet-tasting sucrose solution relative to unsweetened water, we performed the 2-bottle sucrose test as following: Prior to the assay, mice were singly housed and were habituated to the presence of two drinking bottles for 3 days in their cages. At day 4, mice were presented with 2 bearing sipper tubes, one tube containing plain drinking water and the second containing 2% (W/V) sucrose solution (Sigma, S7903–1KG, Lot#SLCJ1863). Mice had the free choice of either drinking the 2% sucrose solution or plain water for a period of 4 days. Then, we repeated the same procedure with 5% sucrose solution for another 4-day period. During the testing, the positions of two bottles were switched daily to reduce the side bias. Consumptions of drinking water and sucrose water were measured daily, and the cages were changed every other day due to the frequent urination over sweet sucrose solution drinking.

Measurement of serum and other tissue ethanol and AcH by gas chromatograph-mass spectrometer (GC–MS):

The mice were anesthetized with isoflurane for the tissue harvest. Left lobes of mouse liver, cerebellar cortex, and intestine tissues (mainly duodenum) were quickly taken after sacrificing the mice at specific timepoints, and all tissue were promptly frozen in liquid nitrogen. All tissues were kept at −80°C for long-term storage. The GC/MS method was essentially performed as prescribed in Jin et al. (2021)¹⁹. All procedures were conducted at 0–4 °C. Briefly, 20–50 mg of frozen tissue or serum/bile (as described above) were prepared before measurement, and were mixed with 5 μmol of ²H₆-EtOH (internal standard for EtOH) and 0.04 μmol of ²H₄-AcH (internal standard for acetaldehyde) prior to adding 250 μl of 0.6 N perchloric acid into each sample. These samples were homogenized at 4°C for 2 mins (Precellys® Evolution Homogenizer, Bertin) and then centrifuged at 13,200g ×15 min at 4°C. The 200 μl supernatant of each sample was quantitatively transferred into a 20 ml headspace vial and capped immediately. Headspace vials were then loaded onto the 108-vial tray of a headspace sampler coupled to gas chromatography-mass spectrometry (GC/MS, Agilent Technologies, Santa Clara, CA). The concentrations of EtOH and AcH in serum and other tissue samples were calculated by comparing the integrated areas of EtOH and AcH peaks on the gas chromatograms with those of internal standards added in each sample¹⁹.

Statistical analyses:

In general, data collection and analysis were not performed blindly to the conditions of the experiments. No animals or data points were excluded from the analysis except for drinking data points that were <0 after weighing and correcting for the blank bottles. Data distribution was assumed to be normal but this was not formally tested. Data were presented as the means±SEM and were analyzed by using GraphPad Prism software (V.9.0; GraphPad Software, La Jolla, California, USA). To compare values from two groups, significance was evaluated by Student t-test, with paired t test for the matched values. Data from multiple groups were compared with one-way ANOVA or two-way ANOVA analysis followed by Tukey’s post hoc test. Pearson correlation analysis was performed to evaluate the linear relationship of two parameters. All statistical tests were two-sided. P values of <0.05 were considered statistically significant.

Extended Data

Extended Data Fig. 3| — **(a)** Western blot analysis was performed to determine the ALDH2 protein expression in liver and different segments of intestine tissues, including duodenum, mid small intestine, and terminal ileum (n=4 in each group) (**p=0.0021,**p=0.0078). **(b)** Representative immunofluorescence images showing expression of ALDH2 in liver and duodenum tissues from WT, *Aldh2* ^Hep−/−, *Aldh2* ^{villin−/−}, and *Aldh2* ^{Hep−/−Villin−/−} mice. **(c)** Schematic of mouse model of chronic ethanol feeding plus acute binge (the NIAAA model) and its pair-fed control model (Created with Biorender.com). **(d)** Serum ALT (IU/L) (liver injury marker) measurements of chronic-plus-binge ethanol feeding mice (shown as ‘EtOH diet’) and pair-fed control mice (shown as ‘Paired diet’) of WT, *Aldh2* ^Hep−/−, *Aldh2* ^{villin−/−}*, Aldh2* ^{Hep−/−Villin−/−} mice (n=4 in each paired fed group; n=13, n=12, n=8, n=7 in each EtOH-fed group), the results were from one experiment. Values represent means±SEM. **p<0.01. Two-sided Student’s t-test and one-way ANOVA were used for the comparison between two groups. ns: No significance.

Extended Data Fig. 4| — **(a)** Representative images of hematoxylin-eosin (H&E) staining and IHC staining with anti-IBA1 (macrophage marker), anti-myeloperoxidase (MPO; neutrophil marker; positive cells were indicated with black arrows), Sirius red (fibrosis), and anti-OPN (bile duct marker) of liver sections from WT, *Aldh2*^Hep−/−*, Aldh2*^{villin−/−}, and *Aldh2*^{Hep−/−Villin−/−} mice of chronic-plus-binge ethanol feeding model (n=5 in each group). **(b)** Quantification of IHC staining shown in panel (a) (count of positive cell or fold change of positive area/200×field) (n=5 in each group) (**p=0.0082, **p=0.0059, **p=0.0022, **p=0.0016). Values represent means±SEM. *p<0.05; **p<0.01. Two-sided Student’s t-test and one-way ANOVA were used for the comparison between indicated two groups.

Extended Data Fig. 5| — **(a)** Measurement of EtOH and AcH in serum and bile samples from C57BL/6N mice (n=5) by GC-MS 1h and 3h post *i.p* injection of ethanol (4g/Kg). Box plot with whiskers (min to max), line at median were shown in (a). Two-sided paired Student’s t-test was performed, **p=0.0028, **p=0.0015. (b) Bile volume/body weight ratios (shown as Bile/BW ratio) of C57BL/6N mice received PBS gavage or ethanol (shown as EtOH gavage; 5g/Kg) were determined 3h, 6h, and 9h post oral gavage (n=4 per group at each timepoint). (c) Schematic of potential AcH metabolite downstream of glutathione (GSH) and cysteinylglycine (CysGly). (d) Study timeline–male C57BL/6N mice (n=7/group) were administered buthionine sulfoximine (BSO, 4mmol/kg) or vehicle (control) 2.5h before 5g/kg EtOH gavage and mice were sacrificed 1h later. (e, f) Measurement of EtOH and AcH in serum, liver (n=7, n=7), and bile samples from C57BL/6N mice (n=4) by GC-MS (**p=0.0098). (g-i). Study timeline of male C57BL/6N mice (n=7/group) were administered BSO was shown in (g), and measurement of EtOH and AcH in serum, liver, and bile samples (n=7/group) were shown in (h) and (i). (d) and (g) were created with Biorender.com. Values represent means±SEM. Significance was evaluated via two-sided unpaired Student’s t-test (**p<0.01). ns: No significance.

Extended Data Fig. 6| — **(a)** Western blot analysis of ALDH2 protein in liver tissues from WT and GF mice (n=6). **(b)** Bile volumes (Left panel) and the fold change of bile volume/body weight ratio (shown as Bile/BW ratio) from WT and GF mice post ethanol gavage (5g/Kg) were determined (n=7) (***p<0.0001). Values represent means±SEM. ***p<0.001. Two-sided Student’s t-test was used for the comparison between indicated two groups.

Extended Data Fig. 7| — **(A)** EtOH levels in liver tissue, serum, bile samples from BDL mice (n=5) and sham mice (n=5) 3h post ethanol (5g/kg) gavage. **(b)** EtOH levels in liver, serum, bile samples from *Mdr2* KO mice (*Mdr2*^−/−) (n=5) and control WT mice (n=5) 3h post ethanol (5g/kg) gavage. **(c)** EtOH levels in liver, serum, bile samples from *Bsep* KO mice (*Bsep*^−/−) (n=6) and control WT mice (n=7) 3h post ethanol (5g/kg) gavage. **(d)** EtOH levels in liver tissue, serum, and bile samples from C57BL/6N mice fed with control diet (n=5) and Ursodeoxycholic acid (UDCA) diet (n=5) 3h post ethanol gavage (5g/Kg). **(e-h)** EtOH levels in liver, serum, and bile samples from C57BL/6N mice pre-treated with vehicle, or Cyclo-1 (n=3, n=4), Novobiocin (n=7, n=4), Quinidine (n=5, n=5) and Rifampicin (n=5, n=5), respectively, were measured 3h post ethanol gavage (5g/Kg) (n=3–6 each group) (*p=0.0159). Values represent means±SEM. Two-sided Student’s t-test was used for the comparison between indicated two groups. *p<0.05.

Extended Data Fig. 8| — **(a, b)** Liver tissues were obtained from bile duct ligated (BDL) (n=6), *Mdr2*^−/− (n=6, **p=0.0082), *Bsep*^−/− (n=5) or UDCA treated mice (n=6), and their corresponding control mice. Western blot analyses were performed to determine ALDH2 expression. **(c)** Comparison of ALDH2 enzymatic activity in fresh liver homogenates was performed in the mice mentioned above (n=6, n=6, n=5, n=6). (**d, e)** DID assay in female C57BL/6N mice fed with chow diet (n=10) or UDCA diet (n=9, *p=0.0464, *p=0.0301), in female *Mdr2* KO mice (*Mdr2*^−/−) (n=7, *p=0.0256, **p=0.0015) and their littermate control mice (WT) (n=8), and in male *Bsep* KO mice (*Bsep*^−/−) (n=7, *p=0.0117, *p=0.0111) and their littermate control mice (WT) (n=7). Values represent means±SEM. Two-sided Student’s t-test was used for the comparison between indicated two groups. *p<0.05, **p<0.01

Extended Data Fig. 9| — **(a)** AcH levels of cerebellar cortex, portal blood, bile and duodenal luminal content from four groups of mice were determined (female groups: n=8, n=8, n=8, n=7) (*p=0.0228,***p<0.0001,***p<0.0001,**p=0.0023,*p=0.0194,***p=0.0004,*p=0.0207, ***p<0.0001,***p<0.0001,**p=0.0043,***p=0.0009,***p<0.0001). **(b)** EtOH in cerebellar cortex, portal blood, bile and duodenal luminal content from four groups of mice were determined (male groups: n=9, n=8, n=9, n=11). **(c)** AcH levels from male WT and double KO mice treated with 2g/kg EtOH gavage (*p=0.0260, **p=0.0087, *p=0.0189, *p=0.0260) (The scheme was created with Biorender.com). **(d)** The correlation of serum and cerebellar AcH levels 3h post ethanol gavage (5g/Kg) (n=9, n=9, n=7, n=11). **(e)** Schematic of different brain regions collected for AcH measurement post EtOH gavage (Created with Biorender.com). **(f)** AcH levels of prefrontal cortex (PFC) (n=7, n=7), hippocampus (n=6, n=7), thalamus (TH) (n=6, n=6), and hypothalamus (HTH) (n=7, n=7) from WT (*Aldh2* ^f/f) and *Aldh2* ^{Hep−/−Villin−/−} mice in (e) were determined 3h post EtOH gavage (5g/Kg) (**p=0.0053, ***p=0.0004). *p<0.05, **p<0.01, ***p<0.001. Values represent means±SEM. Two-sided student’s t-test and one-way ANOVA were used for the comparison between two groups in panels a and b; Two-sided student’s t-test was used for panels c and f. Two-tailed simple linear regression was used to determine the correlation.

Extended Data Fig. 10|: — The following parameters of WT mice (*Aldh2*^f/f), *Aldh2*^Hep−/−, *Aldh2*^{Villin−/−}, and *Aldh2* ^{Hep−/−Villin−/−} mice (n=4/each group) were evaluated by using metabolic chambers after ethanol gavage (5g/Kg): **(a)** Respiratory quotient (**p=0.0077). **(b)** Carbohydrate oxidation (*p=0.0133). **(c)** Cumulative food intake. **(d)** Cumulative water intake. **(e)** Total energy expenditure. **(f)** Fat oxidation. **(g)** Oxygen consumption. **(h)** Ambulatory movements. Values represent means±SEM. A two-way ANOVA was performed for the comparisons among multiple groups, followed by two-sided Student’s t-test between WT mice and *Aldh2* ^{Hep−/−Villin−/−} mice in (a) and (b) at the timepoint of 48h. No adjustments were made for multiple comparisons. *p<0.05, **p<0.01.

Supplementary Material

Supplementary Tables

NIHMS2104873-supplement-Supplementary_Tables.pptx^{(44.1KB, pptx)}

Supplementary Materials

NIHMS2104873-supplement-Supplementary_Materials.pdf^{(468.3KB, pdf)}

Acknowledgements:

The authors thank Dr. David Lovinger (NIAAA, NIH) for critical comments and suggestions during the study. The authors also thank the NIAID Gnotobiotic Animal Facility for providing germ-free mice and Dr. Victor Ling (British Columbia Cancer) for providing Bsep ^−/− mice. This work was supported by the intramural program of NIAAA, NIH (B.G., G.K., P.P.). Dr. Huiping Zhou is the recipient of a Research Career Scientist Award from the Department of Veterans Affairs (IK6BX004477). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Footnotes

Code Availability Statement:

No custom code was used in this manuscript.

Competing Interests Statement: All authors declare no competing interests.

Data Availability Statement:

All data from these studies are contained within this manuscript. Source data are provided with this paper.

References:

1.Cederbaum AI Alcohol metabolism. Clin Liver Dis 16, 667–685 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Goedde HW, et al. Distribution of ADH2 and ALDH2 genotypes in different populations. Human Genetics 88, 344–346 (1992). [DOI] [PubMed] [Google Scholar]
3.Yoshida A, Huang IY & Ikawa M Molecular abnormality of an inactive aldehyde dehydrogenase variant commonly found in Orientals. Proceedings of the National Academy of Sciences 81, 258–261 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Lai C-L, et al. Dominance of the Inactive Asian Variant Over Activity and Protein Contents of Mitochondrial Aldehyde Dehydrogenase 2 in Human Liver. Alcohol: Clinical and Experimental Research 38, 44–50 (2014). [DOI] [PubMed] [Google Scholar]
5.Higuchi S, et al. Aldehyde dehydrogenase genotypes In Japanese alcoholics. The Lancet 343, 741–742 (1994). [DOI] [PubMed] [Google Scholar]
6.Kim D-J, et al. Major genetic components underlying alcoholism in Korean population. Human Molecular Genetics 17, 854–858 (2007). [DOI] [PubMed] [Google Scholar]
7.HUGHES JC & COOK CCH The efficacy of disulfiram: a review of outcome studies. Addiction 92, 381–395 (1997). [PubMed] [Google Scholar]
8.Lanz J, et al. Disulfiram: Mechanisms, Applications, and Challenges. Antibiotics (Basel) 12(2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Jorgensen CH, Pedersen B & Tonnesen H The efficacy of disulfiram for the treatment of alcohol use disorder. Alcohol Clin Exp Res 35, 1749–1758 (2011). [DOI] [PubMed] [Google Scholar]
10.O’Malley SS, et al. Interaction of Ethanol and Oral ANS-6637, a Selective ALDH2 Inhibitor in Males: A Randomized, Double-Blind, Placebo-Controlled, Single-Ascending Dose Cohort Study. Alcohol Clin Exp Res 44, 1885–1895 (2020). [DOI] [PubMed] [Google Scholar]
11.Baan R, et al. Carcinogenicity of alcoholic beverages. The Lancet Oncology 8, 292–293 (2007). [DOI] [PubMed] [Google Scholar]
12.Chang JS, Hsiao J-R & Chen C-H ALDH2 polymorphism and alcohol-related cancers in Asians: a public health perspective. Journal of Biomedical Science 24, 19 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Tsai M-C, et al. Association of Heavy Alcohol Intake and ALDH2 rs671 Polymorphism With Hepatocellular Carcinoma and Mortality in Patients With Hepatitis B Virus–Related Cirrhosis. JAMA Network Open 5, e2223511-e2223511 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Zhang J, et al. The role of aldehyde dehydrogenase 2 in cardiovascular disease. Nature Reviews Cardiology 20, 495–509 (2023). [DOI] [PubMed] [Google Scholar]
15.Chen Y-C, et al. Pharmacokinetic and pharmacodynamic basis for overcoming acetaldehyde-induced adverse reaction in Asian alcoholics, heterozygous for the variant ALDH2*2 gene allele. Pharmacogenetics and Genomics 19, 588–599 (2009). [DOI] [PubMed] [Google Scholar]
16.Baillie M Alcohol and the liver. Gut 12, 222–229 (1971). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Eriksson CJP HEPATIC ACETALDEHYDE METABOLISM DURING ETHANOL OXIDATION. in Alcohol and Aldehyde Metabolizing Systems (eds. Thurman RG, Williamson JR, Drott HR & Chance B) 285–294 (Academic Press, 1977). [Google Scholar]
18.Guillot A, et al. Targeting liver aldehyde dehydrogenase-2 prevents heavy but not moderate alcohol drinking. Proc Natl Acad Sci U S A 116, 25974–25981 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Jin S, et al. Brain ethanol metabolism by astrocytic ALDH2 drives the behavioural effects of ethanol intoxication. Nat Metab 3, 337–351 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Mackowiak B, Fu Y, Maccioni L & Gao B Alcohol-associated liver disease. J Clin Invest 134(2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Park SH, et al. Ethanol and its Nonoxidative Metabolites Promote Acute Liver Injury by Inducing ER Stress, Adipocyte Death, and Lipolysis. Cellular and Molecular Gastroenterology and Hepatology 15, 281–306 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Gissen P & Arias IM Structural and functional hepatocyte polarity and liver disease. J Hepatol 63, 1023–1037 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Wang L & Boyer JL The maintenance and generation of membrane polarity in hepatocytes. Hepatology 39, 892–899 (2004). [DOI] [PubMed] [Google Scholar]
24.Hofmann AF Chemistry and enterohepatic circulation of bile acids. Hepatology 4, 4S–14S (1984). [DOI] [PubMed] [Google Scholar]
25.Tuma DJ, Newman MR, Donohue TM Jr. & Sorrell MF Covalent Binding of Acetaldehyde to Proteins: Participation of Lysine Residues. Alcohol: Clinical and Experimental Research 11, 579–584 (1987). [DOI] [PubMed] [Google Scholar]
26.Niemelä O, Israel Y, Mizoi Y, Fukunaga T & Eriksson CJP Hemoglobin-Acetaldehyde Adducts in Human Volunteers Following Acute Ethanol Ingestion. Alcohol: Clinical and Experimental Research 14, 838–841 (1990). [DOI] [PubMed] [Google Scholar]
27.Sprince H, Parker CM, Smith GG & Gonzales LJ Protection against acetaldehyde toxicity in the rat by L-cysteine, thiamin and L-2-methylthiazolidine-4-carboxylic acid. Agents Actions 4, 125–130 (1974). [DOI] [PubMed] [Google Scholar]
28.Anni H, Pristatsky P & Israel Y Binding of acetaldehyde to a glutathione metabolite: mass spectrometric characterization of an acetaldehyde-cysteinylglycine conjugate. Alcohol Clin Exp Res 27, 1613–1621 (2003). [DOI] [PubMed] [Google Scholar]
29.Kera Y, Kiriyama T & Komura S Conjugation of acetaldehyde with cysteinylglycine, the first metabolite in glutathione breakdown by gamma-glutamyltranspeptidase. Agents Actions 17, 48–52 (1985). [DOI] [PubMed] [Google Scholar]
30.Mårtensson J, Jain A, Frayer W & Meister A Glutathione metabolism in the lung: inhibition of its synthesis leads to lamellar body and mitochondrial defects. Proceedings of the National Academy of Sciences 86, 5296–5300 (1989). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Kera Y, Ohbora Y & Komura S Buthionine sulfoximine inhibition of glutathione biosynthesis enhances hepatic lipid peroxidation in rats during acute ethanol intoxication. Alcohol Alcohol 24, 519–524 (1989). [DOI] [PubMed] [Google Scholar]
32.Martino C, et al. Acetate reprograms gut microbiota during alcohol consumption. Nat Commun 13, 4630 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Seitz HK, et al. Possible role of acetaldehyde in ethanol-related rectal cocarcinogenesis in the rat. Gastroenterology 98, 406–413 (1990). [DOI] [PubMed] [Google Scholar]
34.Tedesco D, et al. Alterations in Intestinal Microbiota Lead to Production of Interleukin 17 by Intrahepatic gammadelta T-Cell Receptor-Positive Cells and Pathogenesis of Cholestatic Liver Disease. Gastroenterology 154, 2178–2193 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Kubitz R, Droge C, Stindt J, Weissenberger K & Haussinger D The bile salt export pump (BSEP) in health and disease. Clin Res Hepatol Gastroenterol 36, 536–553 (2012). [DOI] [PubMed] [Google Scholar]
36.Bodewes FA, et al. Ursodeoxycholate modulates bile flow and bile salt pool independently from the cystic fibrosis transmembrane regulator (Cftr) in mice. Am J Physiol Gastrointest Liver Physiol 302, G1035–1042 (2012). [DOI] [PubMed] [Google Scholar]
37.Deters M, Klabunde T, Kirchner G, Resch K & Kaever V Sirolimus/cyclosporine/tacrolimus interactions on bile flow and biliary excretion of immunosuppressants in a subchronic bile fistula rat model. Br J Pharmacol 136, 604–612 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Sanoh S, et al. Changes in Bile Acid Concentrations after Administration of Ketoconazole or Rifampicin to Chimeric Mice with Humanized Liver. Biol Pharm Bull 42, 1366–1375 (2019). [DOI] [PubMed] [Google Scholar]
39.Lenzen R, Stremmel W & Strohmeyer G Antiarrhythmic drugs impair hepatic uptake and secretory function by different mechanisms in the isolated perfused rat liver. Biochim Biophys Acta 1074, 406–412 (1991). [DOI] [PubMed] [Google Scholar]
40.Abaut AY, Chevanne F & Le Corre P Influence of efflux transporters on liver, bile and brain disposition of amitriptyline in mice. Int J Pharm 378, 80–85 (2009). [DOI] [PubMed] [Google Scholar]
41.Thiele TE, Crabbe JC & Boehm SL 2nd. “Drinking in the Dark” (DID): a simple mouse model of binge-like alcohol intake. Curr Protoc Neurosci 68, 9 49 41–49 49 12 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Deng XS & Deitrich RA Putative role of brain acetaldehyde in ethanol addiction. Curr Drug Abuse Rev 1, 3–8 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Chen CH, Ferreira JC, Gross ER & Mochly-Rosen D Targeting aldehyde dehydrogenase 2: new therapeutic opportunities. Physiol Rev 94, 1–34 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Haass-Koffler CL, Akhlaghi F, Swift RM & Leggio L Altering ethanol pharmacokinetics to treat alcohol use disorder: Can you teach an old dog new tricks? J Psychopharmacol 31, 812–818 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Zhang XYF, Gertsik L, Hanrahan J First-in-human clinical trial of DCR-AUD: Evaluation of siRNA knockdown of hepatic acetaldehyde dehydrogenase 2 using a novel bioanalytic assay. Alcohol Clin Exp Res 46, 39A (2022). [Google Scholar]
46.Correa M, et al. Piecing together the puzzle of acetaldehyde as a neuroactive agent. Neuroscience & Biobehavioral Reviews 36, 404–430 (2012). [DOI] [PubMed] [Google Scholar]
47.Tabakoff B, Anderson RA & Ritzmann RF Brain acetaldehyde after ethanol administration. Biochemical Pharmacology 25, 1305–1309 (1976). [DOI] [PubMed] [Google Scholar]
48.Jamal M, et al. Ethanol and Acetaldehyde After Intraperitoneal Administration to Aldh2-Knockout Mice-Reflection in Blood and Brain Levels. Neurochemical Research 41, 1029–1034 (2016). [DOI] [PubMed] [Google Scholar]
49.Westcott JY, Weiner H, Shultz J & Myers RD In vivo acetaldehyde in the brain of the rat treated with ethanol. Biochemical Pharmacology 29, 411–417 (1980). [DOI] [PubMed] [Google Scholar]
50.Chen Y-C, Yang L-F, Lai C-L & Yin S-J Acetaldehyde Enhances Alcohol Sensitivity and Protects against Alcoholism: Evidence from Alcohol Metabolism in Subjects with Variant ALDH2*2 Gene Allele. Biomolecules 11, 1183 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Hendershot CS, et al. Evaluating a Cognitive Model of ALDH2 and Drinking Behavior. Alcoholism: Clinical and Experimental Research 35, 91–98 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Chen C-H, Cruz LA & Mochly-Rosen D Pharmacological recruitment of aldehyde dehydrogenase 3A1 (ALDH3A1) to assist ALDH2 in acetaldehyde and ethanol metabolism in vivo. Proceedings of the National Academy of Sciences 112, 3074–3079 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Mews P, et al. Alcohol metabolism contributes to brain histone acetylation. Nature 574, 717–721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Stellaard F & Lutjohann D Dynamics of the enterohepatic circulation of bile acids in healthy humans. Am J Physiol Gastrointest Liver Physiol 321, G55–G66 (2021). [DOI] [PubMed] [Google Scholar]
55.Im PK, et al. Alcohol drinking and risks of total and site-specific cancers in China: A 10-year prospective study of 0.5 million adults. International Journal of Cancer 149, 522–534 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Huang J, et al. Updated epidemiology of gastrointestinal cancers in East Asia. Nature Reviews Gastroenterology & Hepatology 20, 271–287 (2023). [DOI] [PubMed] [Google Scholar]
57.Kurtin WE, Schwesinger WH & Stewart RM Effect of dietary ethanol on gallbladder absorption and cholesterol gallstone formation in the prairie dog. The American Journal of Surgery 161, 470–474 (1991). [DOI] [PubMed] [Google Scholar]
58.Leitzmann MF, et al. Prospective Study of Alcohol Consumption Patterns in Relation to Symptomatic Gallstone Disease in Men. Alcohol: Clinical and Experimental Research 23, 835–841 (1999). [PubMed] [Google Scholar]
59.Cha BH, Jang M. j. & Lee SH Alcohol Consumption Can Reduce the Risk of Gallstone Disease: A Systematic Review with a Dose-Response Meta-Analysis of Case-Control and Cohort Studies. Gut and Liver 13, 114–131 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Thus C, Knipschild P & Leffers P Does alcohol protect against the formation of gallstones? a demonstration of protopathic bias. Journal of Clinical Epidemiology 44, 941–946 (1991). [DOI] [PubMed] [Google Scholar]
61.Wang R, et al. Hydrophilic bile acids prevent liver damage caused by lack of biliary phospholipid in Mdr2−/− mice[S]. Journal of Lipid Research 60, 85–97 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Wu X, et al. Satiety induced by bile acids is mediated via vagal afferent pathways. JCI Insight 5(2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Castellanos-Jankiewicz A, et al. Hypothalamic bile acid-TGR5 signaling protects from obesity. Cell Metabolism 33, 1483–1492.e1410 (2021). [DOI] [PubMed] [Google Scholar]
64.Kuhre RE, et al. Bile acids are important direct and indirect regulators of the secretion of appetite- and metabolism-regulating hormones from the gut and pancreas. Molecular Metabolism 11, 84–95 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Matsumoto A, Thompson DC, Chen Y, Kitagawa K & Vasiliou V Roles of defective ALDH2 polymorphism on liver protection and cancer development. Environ Health Prev Med 21, 395–402 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Chen C-C, et al. Interaction between the Functional Polymorphisms of the Alcohol-Metabolism Genes in Protection against Alcoholism. The American Journal of Human Genetics 65, 795–807 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Yokoyama A, et al. Genetic polymorphisms of alcohol and aldehyde dehydrogenases and glutathione S -transferase M1 and drinking, smoking, and diet in Japanese men with esophageal squamous cell carcinoma. Carcinogenesis 23, 1851–1859 (2002). [DOI] [PubMed] [Google Scholar]
68.Brooks PJ, Enoch M-A, Goldman D, Li T-K & Yokoyama A The Alcohol Flushing Response: An Unrecognized Risk Factor for Esophageal Cancer from Alcohol Consumption. PLOS Medicine 6, e1000050 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Seike T, Chen CH & Mochly-Rosen D Impact of common ALDH2 inactivating mutation and alcohol consumption on Alzheimer’s disease. Front Aging Neurosci 15, 1223977 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Luczak SE, et al. Effects of ALDH2∗2 on alcohol problem trajectories of Asian American college students. J Abnorm Psychol 123, 130–140 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Siciliano CA, et al. A cortical-brainstem circuit predicts and governs compulsive alcohol drinking. Science 366, 1008–1012 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Brown AR, et al. Structured tracking of alcohol reinforcement (STAR) for basic and translational alcohol research. Molecular Psychiatry 28, 1585–1598 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Winters ND, et al. Targeting diacylglycerol lipase reduces alcohol consumption in preclinical models. The Journal of Clinical Investigation 131(2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Arolfo MP, et al. Suppression of Heavy Drinking and Alcohol Seeking by a Selective ALDH-2 Inhibitor. Alcoholism: Clinical and Experimental Research 33, 1935–1944 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
75.O’Malley SS, et al. Interaction of Ethanol and Oral ANS-6637, a Selective ALDH2 Inhibitor in Males: A Randomized, Double-Blind, Placebo-Controlled, Single-Ascending Dose Cohort Study. Alcoholism: Clinical and Experimental Research 44, 1885–1895 (2020). [DOI] [PubMed] [Google Scholar]
76.Wang R, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad Sci U S A 98, 2011–2016 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Kelly CJ, et al. Crosstalk between Microbiota-Derived Short-Chain Fatty Acids and Intestinal Epithelial HIF Augments Tissue Barrier Function. Cell Host Microbe 17, 662–671 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Liow JS, et al. Effect of a P-glycoprotein inhibitor, Cyclosporin A, on the disposition in rodent brain and blood of the 5-HT1A receptor radioligand, [11C](R)-(−)-RWAY. Synapse 61, 96–105 (2007). [DOI] [PubMed] [Google Scholar]
79.Asaoka Y, et al. Immunohistochemistry of LAMP-2 and adipophilin for phospholipidosis in liver and kidney in ketoconazole-treated mice. Exp Toxicol Pathol 65, 817–823 (2013). [DOI] [PubMed] [Google Scholar]
80.Fu Y, et al. MicroRNA-223 attenuates hepatocarcinogenesis by blocking hypoxia-driven angiogenesis and immunosuppression. Gut (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Feng D, et al. Monocyte-derived macrophages orchestrate multiple cell-type interactions to repair necrotic liver lesions in disease models. The Journal of Clinical Investigation 133(2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Bertola A, Mathews S, Ki SH, Wang H & Gao B Mouse model of chronic and binge ethanol feeding (the NIAAA model). Nature Protocols 8, 627–637 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Tables

NIHMS2104873-supplement-Supplementary_Tables.pptx^{(44.1KB, pptx)}

Supplementary Materials

NIHMS2104873-supplement-Supplementary_Materials.pdf^{(468.3KB, pdf)}

Data Availability Statement

All data from these studies are contained within this manuscript. Source data are provided with this paper.

[R1] 1.Cederbaum AI Alcohol metabolism. Clin Liver Dis 16, 667–685 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Goedde HW, et al. Distribution of ADH2 and ALDH2 genotypes in different populations. Human Genetics 88, 344–346 (1992). [DOI] [PubMed] [Google Scholar]

[R3] 3.Yoshida A, Huang IY & Ikawa M Molecular abnormality of an inactive aldehyde dehydrogenase variant commonly found in Orientals. Proceedings of the National Academy of Sciences 81, 258–261 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Lai C-L, et al. Dominance of the Inactive Asian Variant Over Activity and Protein Contents of Mitochondrial Aldehyde Dehydrogenase 2 in Human Liver. Alcohol: Clinical and Experimental Research 38, 44–50 (2014). [DOI] [PubMed] [Google Scholar]

[R5] 5.Higuchi S, et al. Aldehyde dehydrogenase genotypes In Japanese alcoholics. The Lancet 343, 741–742 (1994). [DOI] [PubMed] [Google Scholar]

[R6] 6.Kim D-J, et al. Major genetic components underlying alcoholism in Korean population. Human Molecular Genetics 17, 854–858 (2007). [DOI] [PubMed] [Google Scholar]

[R7] 7.HUGHES JC & COOK CCH The efficacy of disulfiram: a review of outcome studies. Addiction 92, 381–395 (1997). [PubMed] [Google Scholar]

[R8] 8.Lanz J, et al. Disulfiram: Mechanisms, Applications, and Challenges. Antibiotics (Basel) 12(2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Jorgensen CH, Pedersen B & Tonnesen H The efficacy of disulfiram for the treatment of alcohol use disorder. Alcohol Clin Exp Res 35, 1749–1758 (2011). [DOI] [PubMed] [Google Scholar]

[R10] 10.O’Malley SS, et al. Interaction of Ethanol and Oral ANS-6637, a Selective ALDH2 Inhibitor in Males: A Randomized, Double-Blind, Placebo-Controlled, Single-Ascending Dose Cohort Study. Alcohol Clin Exp Res 44, 1885–1895 (2020). [DOI] [PubMed] [Google Scholar]

[R11] 11.Baan R, et al. Carcinogenicity of alcoholic beverages. The Lancet Oncology 8, 292–293 (2007). [DOI] [PubMed] [Google Scholar]

[R12] 12.Chang JS, Hsiao J-R & Chen C-H ALDH2 polymorphism and alcohol-related cancers in Asians: a public health perspective. Journal of Biomedical Science 24, 19 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Tsai M-C, et al. Association of Heavy Alcohol Intake and ALDH2 rs671 Polymorphism With Hepatocellular Carcinoma and Mortality in Patients With Hepatitis B Virus–Related Cirrhosis. JAMA Network Open 5, e2223511-e2223511 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Zhang J, et al. The role of aldehyde dehydrogenase 2 in cardiovascular disease. Nature Reviews Cardiology 20, 495–509 (2023). [DOI] [PubMed] [Google Scholar]

[R15] 15.Chen Y-C, et al. Pharmacokinetic and pharmacodynamic basis for overcoming acetaldehyde-induced adverse reaction in Asian alcoholics, heterozygous for the variant ALDH2*2 gene allele. Pharmacogenetics and Genomics 19, 588–599 (2009). [DOI] [PubMed] [Google Scholar]

[R16] 16.Baillie M Alcohol and the liver. Gut 12, 222–229 (1971). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Eriksson CJP HEPATIC ACETALDEHYDE METABOLISM DURING ETHANOL OXIDATION. in Alcohol and Aldehyde Metabolizing Systems (eds. Thurman RG, Williamson JR, Drott HR & Chance B) 285–294 (Academic Press, 1977). [Google Scholar]

[R18] 18.Guillot A, et al. Targeting liver aldehyde dehydrogenase-2 prevents heavy but not moderate alcohol drinking. Proc Natl Acad Sci U S A 116, 25974–25981 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Jin S, et al. Brain ethanol metabolism by astrocytic ALDH2 drives the behavioural effects of ethanol intoxication. Nat Metab 3, 337–351 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Mackowiak B, Fu Y, Maccioni L & Gao B Alcohol-associated liver disease. J Clin Invest 134(2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Park SH, et al. Ethanol and its Nonoxidative Metabolites Promote Acute Liver Injury by Inducing ER Stress, Adipocyte Death, and Lipolysis. Cellular and Molecular Gastroenterology and Hepatology 15, 281–306 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Gissen P & Arias IM Structural and functional hepatocyte polarity and liver disease. J Hepatol 63, 1023–1037 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Wang L & Boyer JL The maintenance and generation of membrane polarity in hepatocytes. Hepatology 39, 892–899 (2004). [DOI] [PubMed] [Google Scholar]

[R24] 24.Hofmann AF Chemistry and enterohepatic circulation of bile acids. Hepatology 4, 4S–14S (1984). [DOI] [PubMed] [Google Scholar]

[R25] 25.Tuma DJ, Newman MR, Donohue TM Jr. & Sorrell MF Covalent Binding of Acetaldehyde to Proteins: Participation of Lysine Residues. Alcohol: Clinical and Experimental Research 11, 579–584 (1987). [DOI] [PubMed] [Google Scholar]

[R26] 26.Niemelä O, Israel Y, Mizoi Y, Fukunaga T & Eriksson CJP Hemoglobin-Acetaldehyde Adducts in Human Volunteers Following Acute Ethanol Ingestion. Alcohol: Clinical and Experimental Research 14, 838–841 (1990). [DOI] [PubMed] [Google Scholar]

[R27] 27.Sprince H, Parker CM, Smith GG & Gonzales LJ Protection against acetaldehyde toxicity in the rat by L-cysteine, thiamin and L-2-methylthiazolidine-4-carboxylic acid. Agents Actions 4, 125–130 (1974). [DOI] [PubMed] [Google Scholar]

[R28] 28.Anni H, Pristatsky P & Israel Y Binding of acetaldehyde to a glutathione metabolite: mass spectrometric characterization of an acetaldehyde-cysteinylglycine conjugate. Alcohol Clin Exp Res 27, 1613–1621 (2003). [DOI] [PubMed] [Google Scholar]

[R29] 29.Kera Y, Kiriyama T & Komura S Conjugation of acetaldehyde with cysteinylglycine, the first metabolite in glutathione breakdown by gamma-glutamyltranspeptidase. Agents Actions 17, 48–52 (1985). [DOI] [PubMed] [Google Scholar]

[R30] 30.Mårtensson J, Jain A, Frayer W & Meister A Glutathione metabolism in the lung: inhibition of its synthesis leads to lamellar body and mitochondrial defects. Proceedings of the National Academy of Sciences 86, 5296–5300 (1989). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Kera Y, Ohbora Y & Komura S Buthionine sulfoximine inhibition of glutathione biosynthesis enhances hepatic lipid peroxidation in rats during acute ethanol intoxication. Alcohol Alcohol 24, 519–524 (1989). [DOI] [PubMed] [Google Scholar]

[R32] 32.Martino C, et al. Acetate reprograms gut microbiota during alcohol consumption. Nat Commun 13, 4630 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Seitz HK, et al. Possible role of acetaldehyde in ethanol-related rectal cocarcinogenesis in the rat. Gastroenterology 98, 406–413 (1990). [DOI] [PubMed] [Google Scholar]

[R34] 34.Tedesco D, et al. Alterations in Intestinal Microbiota Lead to Production of Interleukin 17 by Intrahepatic gammadelta T-Cell Receptor-Positive Cells and Pathogenesis of Cholestatic Liver Disease. Gastroenterology 154, 2178–2193 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Kubitz R, Droge C, Stindt J, Weissenberger K & Haussinger D The bile salt export pump (BSEP) in health and disease. Clin Res Hepatol Gastroenterol 36, 536–553 (2012). [DOI] [PubMed] [Google Scholar]

[R36] 36.Bodewes FA, et al. Ursodeoxycholate modulates bile flow and bile salt pool independently from the cystic fibrosis transmembrane regulator (Cftr) in mice. Am J Physiol Gastrointest Liver Physiol 302, G1035–1042 (2012). [DOI] [PubMed] [Google Scholar]

[R37] 37.Deters M, Klabunde T, Kirchner G, Resch K & Kaever V Sirolimus/cyclosporine/tacrolimus interactions on bile flow and biliary excretion of immunosuppressants in a subchronic bile fistula rat model. Br J Pharmacol 136, 604–612 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Sanoh S, et al. Changes in Bile Acid Concentrations after Administration of Ketoconazole or Rifampicin to Chimeric Mice with Humanized Liver. Biol Pharm Bull 42, 1366–1375 (2019). [DOI] [PubMed] [Google Scholar]

[R39] 39.Lenzen R, Stremmel W & Strohmeyer G Antiarrhythmic drugs impair hepatic uptake and secretory function by different mechanisms in the isolated perfused rat liver. Biochim Biophys Acta 1074, 406–412 (1991). [DOI] [PubMed] [Google Scholar]

[R40] 40.Abaut AY, Chevanne F & Le Corre P Influence of efflux transporters on liver, bile and brain disposition of amitriptyline in mice. Int J Pharm 378, 80–85 (2009). [DOI] [PubMed] [Google Scholar]

[R41] 41.Thiele TE, Crabbe JC & Boehm SL 2nd. “Drinking in the Dark” (DID): a simple mouse model of binge-like alcohol intake. Curr Protoc Neurosci 68, 9 49 41–49 49 12 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Deng XS & Deitrich RA Putative role of brain acetaldehyde in ethanol addiction. Curr Drug Abuse Rev 1, 3–8 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Chen CH, Ferreira JC, Gross ER & Mochly-Rosen D Targeting aldehyde dehydrogenase 2: new therapeutic opportunities. Physiol Rev 94, 1–34 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Haass-Koffler CL, Akhlaghi F, Swift RM & Leggio L Altering ethanol pharmacokinetics to treat alcohol use disorder: Can you teach an old dog new tricks? J Psychopharmacol 31, 812–818 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Zhang XYF, Gertsik L, Hanrahan J First-in-human clinical trial of DCR-AUD: Evaluation of siRNA knockdown of hepatic acetaldehyde dehydrogenase 2 using a novel bioanalytic assay. Alcohol Clin Exp Res 46, 39A (2022). [Google Scholar]

[R46] 46.Correa M, et al. Piecing together the puzzle of acetaldehyde as a neuroactive agent. Neuroscience & Biobehavioral Reviews 36, 404–430 (2012). [DOI] [PubMed] [Google Scholar]

[R47] 47.Tabakoff B, Anderson RA & Ritzmann RF Brain acetaldehyde after ethanol administration. Biochemical Pharmacology 25, 1305–1309 (1976). [DOI] [PubMed] [Google Scholar]

[R48] 48.Jamal M, et al. Ethanol and Acetaldehyde After Intraperitoneal Administration to Aldh2-Knockout Mice-Reflection in Blood and Brain Levels. Neurochemical Research 41, 1029–1034 (2016). [DOI] [PubMed] [Google Scholar]

[R49] 49.Westcott JY, Weiner H, Shultz J & Myers RD In vivo acetaldehyde in the brain of the rat treated with ethanol. Biochemical Pharmacology 29, 411–417 (1980). [DOI] [PubMed] [Google Scholar]

[R50] 50.Chen Y-C, Yang L-F, Lai C-L & Yin S-J Acetaldehyde Enhances Alcohol Sensitivity and Protects against Alcoholism: Evidence from Alcohol Metabolism in Subjects with Variant ALDH2*2 Gene Allele. Biomolecules 11, 1183 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] 51.Hendershot CS, et al. Evaluating a Cognitive Model of ALDH2 and Drinking Behavior. Alcoholism: Clinical and Experimental Research 35, 91–98 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Chen C-H, Cruz LA & Mochly-Rosen D Pharmacological recruitment of aldehyde dehydrogenase 3A1 (ALDH3A1) to assist ALDH2 in acetaldehyde and ethanol metabolism in vivo. Proceedings of the National Academy of Sciences 112, 3074–3079 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Mews P, et al. Alcohol metabolism contributes to brain histone acetylation. Nature 574, 717–721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R54] 54.Stellaard F & Lutjohann D Dynamics of the enterohepatic circulation of bile acids in healthy humans. Am J Physiol Gastrointest Liver Physiol 321, G55–G66 (2021). [DOI] [PubMed] [Google Scholar]

[R55] 55.Im PK, et al. Alcohol drinking and risks of total and site-specific cancers in China: A 10-year prospective study of 0.5 million adults. International Journal of Cancer 149, 522–534 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.Huang J, et al. Updated epidemiology of gastrointestinal cancers in East Asia. Nature Reviews Gastroenterology & Hepatology 20, 271–287 (2023). [DOI] [PubMed] [Google Scholar]

[R57] 57.Kurtin WE, Schwesinger WH & Stewart RM Effect of dietary ethanol on gallbladder absorption and cholesterol gallstone formation in the prairie dog. The American Journal of Surgery 161, 470–474 (1991). [DOI] [PubMed] [Google Scholar]

[R58] 58.Leitzmann MF, et al. Prospective Study of Alcohol Consumption Patterns in Relation to Symptomatic Gallstone Disease in Men. Alcohol: Clinical and Experimental Research 23, 835–841 (1999). [PubMed] [Google Scholar]

[R59] 59.Cha BH, Jang M. j. & Lee SH Alcohol Consumption Can Reduce the Risk of Gallstone Disease: A Systematic Review with a Dose-Response Meta-Analysis of Case-Control and Cohort Studies. Gut and Liver 13, 114–131 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Thus C, Knipschild P & Leffers P Does alcohol protect against the formation of gallstones? a demonstration of protopathic bias. Journal of Clinical Epidemiology 44, 941–946 (1991). [DOI] [PubMed] [Google Scholar]

[R61] 61.Wang R, et al. Hydrophilic bile acids prevent liver damage caused by lack of biliary phospholipid in Mdr2−/− mice[S]. Journal of Lipid Research 60, 85–97 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] 62.Wu X, et al. Satiety induced by bile acids is mediated via vagal afferent pathways. JCI Insight 5(2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Castellanos-Jankiewicz A, et al. Hypothalamic bile acid-TGR5 signaling protects from obesity. Cell Metabolism 33, 1483–1492.e1410 (2021). [DOI] [PubMed] [Google Scholar]

[R64] 64.Kuhre RE, et al. Bile acids are important direct and indirect regulators of the secretion of appetite- and metabolism-regulating hormones from the gut and pancreas. Molecular Metabolism 11, 84–95 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R65] 65.Matsumoto A, Thompson DC, Chen Y, Kitagawa K & Vasiliou V Roles of defective ALDH2 polymorphism on liver protection and cancer development. Environ Health Prev Med 21, 395–402 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R66] 66.Chen C-C, et al. Interaction between the Functional Polymorphisms of the Alcohol-Metabolism Genes in Protection against Alcoholism. The American Journal of Human Genetics 65, 795–807 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R67] 67.Yokoyama A, et al. Genetic polymorphisms of alcohol and aldehyde dehydrogenases and glutathione S -transferase M1 and drinking, smoking, and diet in Japanese men with esophageal squamous cell carcinoma. Carcinogenesis 23, 1851–1859 (2002). [DOI] [PubMed] [Google Scholar]

[R68] 68.Brooks PJ, Enoch M-A, Goldman D, Li T-K & Yokoyama A The Alcohol Flushing Response: An Unrecognized Risk Factor for Esophageal Cancer from Alcohol Consumption. PLOS Medicine 6, e1000050 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R69] 69.Seike T, Chen CH & Mochly-Rosen D Impact of common ALDH2 inactivating mutation and alcohol consumption on Alzheimer’s disease. Front Aging Neurosci 15, 1223977 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R70] 70.Luczak SE, et al. Effects of ALDH2∗2 on alcohol problem trajectories of Asian American college students. J Abnorm Psychol 123, 130–140 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R71] 71.Siciliano CA, et al. A cortical-brainstem circuit predicts and governs compulsive alcohol drinking. Science 366, 1008–1012 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R72] 72.Brown AR, et al. Structured tracking of alcohol reinforcement (STAR) for basic and translational alcohol research. Molecular Psychiatry 28, 1585–1598 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R73] 73.Winters ND, et al. Targeting diacylglycerol lipase reduces alcohol consumption in preclinical models. The Journal of Clinical Investigation 131(2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R74] 74.Arolfo MP, et al. Suppression of Heavy Drinking and Alcohol Seeking by a Selective ALDH-2 Inhibitor. Alcoholism: Clinical and Experimental Research 33, 1935–1944 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R75] 75.O’Malley SS, et al. Interaction of Ethanol and Oral ANS-6637, a Selective ALDH2 Inhibitor in Males: A Randomized, Double-Blind, Placebo-Controlled, Single-Ascending Dose Cohort Study. Alcoholism: Clinical and Experimental Research 44, 1885–1895 (2020). [DOI] [PubMed] [Google Scholar]

[R76] 76.Wang R, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad Sci U S A 98, 2011–2016 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] 77.Kelly CJ, et al. Crosstalk between Microbiota-Derived Short-Chain Fatty Acids and Intestinal Epithelial HIF Augments Tissue Barrier Function. Cell Host Microbe 17, 662–671 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R78] 78.Liow JS, et al. Effect of a P-glycoprotein inhibitor, Cyclosporin A, on the disposition in rodent brain and blood of the 5-HT1A receptor radioligand, [11C](R)-(−)-RWAY. Synapse 61, 96–105 (2007). [DOI] [PubMed] [Google Scholar]

[R79] 79.Asaoka Y, et al. Immunohistochemistry of LAMP-2 and adipophilin for phospholipidosis in liver and kidney in ketoconazole-treated mice. Exp Toxicol Pathol 65, 817–823 (2013). [DOI] [PubMed] [Google Scholar]

[R80] 80.Fu Y, et al. MicroRNA-223 attenuates hepatocarcinogenesis by blocking hypoxia-driven angiogenesis and immunosuppression. Gut (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R81] 81.Feng D, et al. Monocyte-derived macrophages orchestrate multiple cell-type interactions to repair necrotic liver lesions in disease models. The Journal of Clinical Investigation 133(2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R82] 82.Bertola A, Mathews S, Ki SH, Wang H & Gao B Mouse model of chronic and binge ethanol feeding (the NIAAA model). Nature Protocols 8, 627–637 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Coordinated action of a gut-liver pathway drives alcohol detoxification and consumption

Yaojie Fu

Bryan Mackowiak

Yuhong Lin

Luca Maccioni

Taylor Lehner

Hongna Pan

Yukun Guan

Grzegorz Godlewski

Hongkun Lu

Cheng Chen

Shoupeng Wei

Dechun Feng

Janos Paloczi

Huiping Zhou

Pal Pacher

Li Zhang

George Kunos

Bin Gao

Abstract

Introduction

Results

Liver and gut ALDH2 synergistically clear blood AcH

Fig. 1: Liver and gut synergistically promote circulating acetaldehyde clearance via the ALDH2.

Bile efflux mediates AcH secretion from liver into gut lumen

Fig. 2: Bile flow is an important pathway for AcH clearance from liver into intestinal lumen.

Gut AcH clearance is not affected by gut microbiota

Fig. 3: Gut microbiota play a minor role in gut luminal AcH clearance.

Gut AcH is cleared via gut/liver ALDH2 first-pass metabolism

Fig. 4: Gut AcH is metabolized by gut ALDH2 with a small portion absorbed back and metabolized by the liver ALDH2.

Bile flow dynamics regulate blood and liver AcH clearance

Fig. 5: Bile flow controls systemic and liver AcH detoxification.

Bile flow dynamics control drinking behavior

Fig. 6: Manipulating intrahepatic bile flow affects drinking behavior.

Liver and gut ALDH2 synergistically promote alcohol drinking

Fig. 7: Gut and liver ALDH2 synergistically control blood acetaldehyde clearance and drinking behavior.

Discussion:

Fig. 8: The schematic for a liver-gut loop controlling alcohol detoxification and drinking behavior.

Methods:

Animals:

Acute ethanol gavage:

Acute ethanol intraperitoneal injection:

Treatment of antibiotics:

UDCA treatment:

Bile/serum sample collection and preparation for EtOH/AcH measurement:

Measurement of AcH/EtOH in intestinal luminal content:

Treatment with different bile transport inhibitors:

Bile duct ligation (BDL):

In vivo intra-intestinal AcH solution injection:

Mouse liver ethanol perfusion:

Western blot:

Immunofluorescent staining:

Liver ALDH2 enzymatic activity measurement:

Mouse model of chronic ethanol diet feeding

RNA isolation and real-time quantitative PCR (RT-qPCR)

Immunohistochemistry (IHC) staining

Behavioral studies:

(1). Dynamic measurement of metabolic parameters:

(2). Drinking in the dark (DID) assay:

(3). Two-bottle choice (2BC) assay:

(4). Rotarod Test

(5). Sucrose preference test

Measurement of serum and other tissue ethanol and AcH by gas chromatograph-mass spectrometer (GC–MS):

Statistical analyses:

Extended Data

Extended Data Fig. 1|. Confirmation of ALDH2 deletion in various strains of single organ Aldh2 KO mice.

Extended Data Fig. 2. Drinking behavior in various strains of single organ Aldh2 KO mice.

Extended Data Fig. 3|. Expression of ALDH2 protein in the gut, and liver injury in gut-specific Aldh2 KO mice post ethanol feeding.

Extended Data Fig. 4|. Measurement of liver injury in liver and/or gut Aldh2 KO post chronic-plus-binge ethanol feeding.

Extended Data Fig. 5|. Bile AcH levels are much higher than serum AcH after ethanol administration and effects of glutathione depletion on acetaldehyde disposition.

Extended Data Fig. 6|. No differences on ALDH2 protein levels between WT and GF mice.

Extended Data Fig. 7|. Manipulation of intrahepatic bile flow does not affect EtOH transportation.

Extended Data Fig. 8|. Expression and enzymatic activity of liver ALDH2 and DID in mice with various treatment or gene deletion.

Extended Data Fig. 9|. ALDH2 of liver-gut loop controls AcH clearance but does not affect EtOH concentration.

Extended Data Fig. 10|: Liver and gut epithelium Aldh2 double knockout leads to significant inhibition of metabolic phenotypes after alcohol intake.

Supplementary Material

Acknowledgements:

Footnotes

Data Availability Statement: