Activating GLP-1 receptors reduces the liver's breakdown of alcohol

Animal studies were conducted according to guidelines set by the National Institutes of Health (NIH) and protocols approved by the Animal Care and Use Committee of Yale University. Mice were kept in individual ventilated cages.

First experiment: (Ethanol study), twenty-four male C57BL/6 N mice aged 12–14 weeks were purchased from NIAAA and were kept in an environment with controlled temperature (23˚C ± 1 °C) and a 12-h light/12-h dark cycle. They were given unrestricted access to both water and standard mouse chow after a pre-acclimation period of one week. Subsequently, the mice were acclimated to the Lieber-Decarli liquid control diet for five days, ad libitum. Half of the cohort were continued on the control diet while the other half were switched to an ethanol diet. Within both groups, mice were subdivided to receive subcutaneous semaglutide or saline for 15 days (Supplementary Fig. 1A). Thus, the groups were as follows: Group 1 (n = 4) - control liquid diet; Group 2 (n = 8) - ethanol diet; Group 3 (n = 8) - ethanol diet with subcutaneous semaglutide; and Group 4 (n = 4) - control diet with subcutaneous semaglutide.

Second experiment: (Fasting study), We obtained male and female C57BL/6 N mice aged 12–14 weeks, and they were given unrestricted access to both water and standard mouse chow after a pre-acclimatizing period of 1 week. The mice were separated into four groups. Group 1 (n = 4) – chow diet; Group 2 (n = 5) – chow diet with semaglutide; Group 3 (n = 8) 16-h fasting mice; and Group 4 (n = 7) 16-h fasting mice with semaglutide. The mice received a single dose of 30 nmol/kg of subcutaneous injection of semaglutide before fasting. After 16 h mice were euthanized, and liver tissues were collected.

A priori, it was established that any animal showing signs of illness or distress prior to the start of the experiment would be excluded from participation. Criteria for exclusion included weight loss greater than 10% of the group mean, reduced food or water intake, or lack of grooming behavior. No animals met these criteria, and thus none were excluded. The experimental procedures were performed by regulations adopted by the National Institute of Health and approved by the Institutional Animal Care and Use Committee, Yale University. Anesthesia was induced using isoflurane, and animals were sacrificed by cervical dislocation under isoflurane anesthesia.

Chronic plus binge ethanol feeding was given as per the NIAAA protocol¹⁶. The Lieber-Decarli control liquid diet was prepared by combining 225 g of dry mix (Bio-Serv, Product F1259SP) with 860 ml of potable water to produce 1000 ml of control liquid diet. The ethanol liquid diet consisted of 133 g of dry mix (Bio-Serv, Product F1258SP), 20.3 g of maltose dextrin, 910 ml of potable water, and 52.6 ml of 95% ethanol, producing 1000 ml of ethanol liquid diet. For the maltose gavage, 9 g of maltose dextrin was dissolved in 20 ml of water. The ethanol gavage was prepared by mixing 6.6 ml of 95% ethanol with 13.4 ml of water to produce 20 ml of solution. Mice received gavage once on day 16. Body weight and food consumption were tracked daily. For the second experiment (Fasting study) mice on the chow arm were maintained on a standard chow diet, While the fasting arm mice have only access to the water for 16 h.

For the first experiment (ethanol study), mice in Groups 3 and 4 (semaglutide groups) received semaglutide (OZEMPIC; NDC Code: 0169-4130-01) via subcutaneous injection every morning at 9:00 am for a period of 15 days post liquid diet acclimatization¹². Semaglutide dose escalation was performed to a target dose of 30 nmol/kg as follows: 1 nmol/kg on days 1 and 2; 3 nmol/kg on days 3 and 4; 10 nmol/kg on days 5 and 6; and 30 nmol/kg on days 7 through 15. Semaglutide doses were based on published data¹⁷. The dose titration was performed to minimize GLP-1A-associated side effects. The other two groups were administered 0.9% sodium chloride injections to serve as a control for any potential stress due to the injection.

For the second experiment (Fasting study) a subgroup from the fasting arm and a subgroup from the chow arm received either single dose of semaglutide 30 nmol/kg or 0.9% sodium chloride injections to serve as control for stress due to injection.

For the first experiment (ethanol study), on day 16, mice in Groups 2 and 3 (Lieber-Decarli ethanol diet groups) were gavaged with 5 g/kg of body weight ethanol, while mice in Groups 1 and 4 were gavaged with 9 g/kg of body weight maltose dextrin. At 9 h following gavage, mice were euthanized and blood samples and tissue samples were collected for histology, RNA and protein analyses. For histology, the left hepatic lobe was enclosed in a prelabelled histology cassette and fixed in 10% neutral buffered formalin for 24 h, followed by 70% ethanol for another 24 h before submission for tissue processing. Other portions of the liver were snap-frozen in liquid nitrogen until further processing.

For the second experiment (Fasting study): Mice were sacrificed after 16 h of fasting, and livers were collected and snap-frozen in liquid nitrogen until further processing.

Formalin-fixed liver tissues were processed for paraffin embedding and stained with hematoxylin and eosin. The slides were scanned using a Motic Easy Scan slide scanner at 40× objective setting and were assessed using the Motic DS assistant software. Paraffin sections were immunostained to detect PLIN2, 4HNE, and CYP2E1 expression. Briefly, the paraffin sections were cleared in xylene and hydrated in descending grades of ethanol. The slides were subjected to antigen retrieval via citrate buffer pH 6.5 (Vector Labs) using a heat retrieval protocol. The slides were then blocked with 2% goat serum in 1 × TBST for 1 h at room temperature, and incubated with anti-mouse PLIN2 antibody (Thermo Fisher Scientific Cat #15294-1AP), anti-mouse 4HNE antibody (Thermo Fisher Scientific Cat #MA5-27570) and anti-rabbit CYP2E1 antibody (Abcam Cat #ab28146). The recommended dilution was used overnight at 4 °C followed by staining with anti-rabbit/mouse antibody tagged with Alexa 555 with recommended dilution for 1 h at room temperature. The slides were washed and mounted with Prolong gold anti-fade mountant with DAPI (Thermo Fisher Scientific) and imaged using white light laser Leica SP8 confocal laser scanning microscope. The images were mostly acquired under a 40 × oil immersion objective with 1024 × 1024 pixel dimensions. Photomultiplier tube detectors were used for abundant proteins at visible range wavelengths. The images from immunostaining were exported as Leica image files and were processed using LASX software. The images were uniformly amended with minimal Gaussian blur to reduce graininess.

Fluorescence images of immunostained liver tissue sections were taken with a WLL (while light laser) Leica Stellaris DIVE confocal microscope. The images were mostly acquired under a 25× water immersion objective with 1024 × 1024 pixel dimensions. A highly sensitive HyD detector with gating was used for imaging. The images from immunostaining were exported as Leica image files (LIF) and were processed using IMARIS software version 9.8 (bitplane). The images were uniformly amended with minimal Gaussian blur to reduce graininess. The images of individual channels and/or overlay were saved as high-resolution TIF files. Further, the exported images were analyzed using the ImageJ software package (NIH) for further analysis.

RNA was extracted from snap-frozen mouse liver samples. Approximately 10 mg of liver tissue was homogenized in Qiazol lysis buffer, and RNA was extracted using MiRNeasy Micro Kit (50) (QIAGEN Cat #74104). Following this, the RNA was transcribed into cDNA employing the oligo(dT) primer and the High-capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific Cat #4368814). Subsequently, quantitative RT-PCR was performed utilizing the Quant Studio 6 flex instrument (Thermo Fisher Scientific Cat #4485691). The expression of each gene of interest (Supplementary Table) was assessed using the Comparative CT method employing SYBR green master mix (Invitrogen Cat #4309155). The CT values of all the groups for expression of each gene of interest were normalized to their respective β-actin values and were presented as relative fold change of gene expression as compared to the control group. 1

The total protein extracts were prepared using RIPA buffer (50 mM Tris, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM Na3VO4, 5 mM NaF, and 1% cocktail protein protease inhibitors (Sigma). Samples were processed on ice and then were centrifuged at 13,200×g for 10 min at 4 °C and 1X protease inhibitor in a microcentrifuge. The concentrations of protein extracts were determined by using the BCA Protein Assay Kit (Thermo Fisher Cat #23225), then protein samples were mixed with 3× loading buffer and loaded in equal amounts into the wells of 4–12% Bis-Tris protein gels (Bio-Rad) to transfer the proteins to PVDF membranes (Millipore).

The primary antibodies used for analysis included anti-CYP2E1 (1:1000; Abcam, Cat #ab28146), anti-β-Actin (1:1000; Cell Signaling, Cat # 4967), and anti-GAPDH (1:1000; Cell Signaling Technology, Cat #2118). Secondary horseradish peroxidase-conjugated antibodies (1:10,000; Cell Signaling Technology, Cat #7076 (mouse) and 7074 (rabbit) were used for analysis.

RNA quality was determined by RNA integrity number with Agilent Technologies 2100 Bioanalyzer (Cat #5067-1511), to confirm that all samples had an RNA integrity number exceeding 9 and were deemed suitable for further analysis. Total RNA (1 μg) was used for library synthesis using TruSeq V2 RNA-Seq kit. RNA sequencing library was prepared via the rRNA depletion method for the first experiment (ethanol study). Quality control analysis and quantification of the sequencing library were performed using Agilent Technologies 2100 Bioanalyzer High Sensitivity DNA Chip (Cat #5067-5592). Paired-end sequencing was performed on Illumina's NovaSeq 6000 sequencing system. The library preparation and sequence analysis were performed by the Yale Center for Genome Analysis (YCGA), and our lab performed bioinformatics analysis.

Quality check and alignment of the reads was done by Partek Flow; HISAT software was used to compare the reads against the mouse reference genome Mus musculus (mouse) – mm39; and mapped reads were obtained using HTSeq. The count matrix was loaded into RStudio (version 4.3.1), prefiltering of low reads (≤10 reads) was performed, and DESeq2 package version 1.42.0 was used to perform statistical analysis and normalize the DEGs between the groups, genes reaching statistical significance with adjusted P value <0.05 between the groups were examined. Over-representation analysis of the DEGs was done by using ShinyGo 0.80 to show the top 20 pathways, ranked based on fold enrichment, and false discovery rate (FDR ≤0.05).

We obtained male and female C57BL/6 N mice aged 12–14 weeks, and they were given unrestricted access to both water and standard mouse chow after a pre-acclimatizing period of 5 days. The mice were separated into two groups; one group was given a single subcutaneous injection of 30 nmol/kg b.wt. semaglutide or subcutaneous saline overnight. On the next day, mice were given a single dose of oral gavage of 2.5 g/kg b.wt. ethanol with a 20 G gavage needle (catalog no. AFN2038C) to both groups. Two additional groups were also injected intraperitoneal (IP) 2.5 g/kg b.wt. ethanol directly into the lower left area of the mouse abdomen with 25G needle and 1-ml syringe to eliminate the decreased gastric emptying effect of GLP-1RA. The experiments were conducted in three parts:

Experiment 1 (male mice): Ethanol (gavage) + saline group (n = 5) vs. ethanol(gavage) + GLP-1R agonist group (n = 4). One animal was excluded due to cessation of breath and death following alcohol gavage.

Experiment 2 (female mice): Ethanol (gavage) + saline group (n = 6) vs. ethanol(gavage) + GLP-1R agonist group (n = 6).

Experiment 3 (male mice, IP ethanol): Ethanol (IP) + saline group (n = 6) vs. ethanol (IP) + GLP-1R agonist group (n = 6) was done on male mice.

Whole blood samples were collected from the tail vein in 20-μl heparinized capillary tubes (Fisher Scientific company LLC #22-362566), anesthesia was induced using isoflurane, the blood samples were transferred quickly (10 s) to 2-ml gas chromatographic septum sealed vials (Sigma #29381-U) containing 200 μl of an internal standard solution consisting of 0.6 N perchloric acid and 4 mM n-propanol in double distilled water on ice. Vials were stored at 4 °C and analyzed the next day by headspace gas chromatography¹⁸.

Headspace analysis was performed by Yale Chemical and Biophysical Instrumentation Center (CBIC). The concentrations of EtOH samples were calculated by comparing the integrated areas of EtOH peaks (3 peaks) on the gas chromatograms with those of 1-propanol peaks (2 peaks), internal standards added in each sample and were normalized to the internal standard (1-propanol). A calibration curve was prepared with known EtOH concentrations and 1-propanol as the internal standard to calculate the absolute ethanol concentration in the blood samples.

A randomization algorithm was used for randomly assigning each cage to the various experimental groups, while treatments were administered in a randomized order per cage. All cages were kept in the same row of each rack. To minimize bias, blinding was applied at various stages of the study. FZ was blinded during the allocation of the groups, conduct of the experiments, and outcome assessments, while AS, WM, and BB were not. All investigators were involved in the data analysis stage were not blinded during this phase. The sample size for the experiments was based on previous experience in the lab with a similar model.

All statistical analyses were performed using Prism 10.0.2 by GraphPad Software, Inc. (CA, USA). Data on ethanol consumption and body weight were analyzed using one-way ANOVA, followed by Šídák's multiple comparisons. qPCR results from the first experiment (ethanol study) were analyzed using one-way ANOVA, followed by Dunnett's test. Immunofluorescent quantifications were analyzed by Two-way ANOVA, followed by post-hoc test (Holm–Šídák's). qPCR and western blot results from the Second experiment (Fasting study) were analyzed using the Mann–Whitney U-test. Western blot results from the first experiment (ethanol study) were analyzed by Brown–Forsythe and Welch's ANOVA tests.

Two-way repeated measures ANOVA was used to compare blood ethanol concentration between the two groups over the five-time points, considering group and time as factors. Follow up with post-hoc (Holm–Šídák's). Analyses were done separately for each experiment based on sex and route of drug administration.