Short-term heavy drinking in a non-human primate model skews monocytes toward a hypo-inflammatory phenotype

Abstract

Introduction: Alcohol use is prevalent in the United States (US), with ~80% of persons over 12 years old reporting alcohol consumption in 2023 and ~10% of those individuals developing alcohol use disorder (AUD). Acute and chronic alcohol consumption exert opposite effects on the immune system. Specifically, acute alcohol exposure (AAE), (3-16 hours of in-vitro treatment, one binge episode in humans, or one gavage feeding in mice) skews monocytes towards a hypo-inflammatory phenotype associated with reduced TNFα, IL-6, and MCP-1 production. In contrast, chronic alcohol consumption (CAC) (7 days of in-vitro treatment, 3-12 months of consumption in animal models, or humans with confirmed AUD diagnosis), shifts the functional, transcriptional, metabolic, and epigenetic landscapes of monocytes and their progenitors towards a hyper-inflammatory profile. Despite the extensive work investigating AAE and CAC, few studies have examined short-term drinking durations. We sought to bridge this gap by assessing monocytes after 6 months of ethanol consumption in a rhesus macaque model, which we considered short-term drinking. Understanding the longitudinal changes in monocytes' phenotype and function in the context of alcohol consumption could pave the way to identifying diagnostic biomarkers for disease progression.

Methods: To bridge this gap, we obtained peripheral blood mononucleated cells (PBMC) isolated from rhesus macaques before and after 6 months of daily ethanol consumption (>55% of intakes over 2.0 g/kg/day). Monocytes were analyzed using a combination of flow cytometry, single-cell RNA-sequencing (scRNAseq), ELISAs, and Cleavage Under Targets and Tagmentation (CUT&Tag).

Results: Our data show that 6 months of ethanol consumption rewires monocytes towards a hypo-inflammatory profile as evidenced by reduced cytokine production. scRNAseq analysis revealed distinct shifts in monocyte states/clusters with ethanol consumption and LPS stimulation in line with a shift to a hypo-inflammatory state. These changes may be driven by reduced levels of H3k4me3, a histone modification shown to be deposited at promoter regions of genes involved in inflammation and pathogen response signaling.

Discussion: Overall, these data demonstrate that 6 months of daily heavy drinking attenuates inflammatory responses in monocytes via shifts in the epigenetic landscape.

Keywords: alcohol; epigenome; hypo-inflammatory; monocytes; non-human primate; transcriptome.

Figure 1

Circulating monocytes display a hypo-inflammatory phenotype after 6mo of heavy ethanol consumption. (A) Study design. (B) Representative gating strategy used to identify and phenotype monocytes within peripheral blood mononuclear cells (PBMC). Bar plots indicating the (C) change in the overall percentage of monocytes, (D) percentage of nonclassical monocytes within the monocyte population, and (E) percentage of CD163+ monocytes and (F) their respective MFI. Statistical significance was determined using a linear mixed model, and error bars were defined as ± standard error of the mean (SEM). A p-value of <0.05 was considered significant while a value 0.1>x>0.5 was denoted as trending.

Figure 2

Reduced IL6 production and phagocytosis in circulating monocytes after 6 months of chronic ethanol consumption. (A) Gating strategy used to identify TNFα+ and IL6+ monocytes. (B) Bar blots comparing the corrected percentage of TNFα+IL6+ monocytes before and after 6mo of heavy ethanol consumption. MFI of (C) IL6 and (D) TNFα from TNFα+IL6+ monocytes after stimulation with bacterial ligands. Bar plot representing the percentage of TNFα+IL+ (E) CD14^high and (F) CD14^low monocytes. (G) Representative gating strategy used to identify Phrodo Red positive (phrodo+) monocytes. (H) Bar plots indicating the percentage of phrodo+ monocytes. Statistical significance was determined using a linear mixed model, and error bars were defined as ± standard error of the mean (SEM).

Figure 3

Heavy ethanol consumption for 6 months and LPS stimulation drive distinct monocyte sub-populations. (A) Uniform Manifold Approximation and Projection (UMAP) of 37,690 CD14+ monocytes. (B) UMAPS displaying the contribution of cells from respective time points/stimulation conditions for each cluster. (C) Selected marker genes used to identify clusters. (D) Functional enrichment of all marker genes by cluster. (E) Relative abundance of clusters stratified by timepoint/stimulation condition (* indicates differences between BL and 6mo; # represents differences between NS and LPS stimulation). A one-way ANOVA between unmatched samples with Šídák correction for multiple comparisons was used to determine differences in cell cluster frequencies across each sample. All bar plots depict the mean ± SEM. A p-value of <0.05 was considered significant. ** p-value <0.0021; *** p-value <0.0002; **** p-value <0.0001; #### p-value <0.0001.

Figure 4

Altered monocyte transcriptional profiles after 6 months of chronic ethanol consumption. (A) Volcano plot depicting genes differentially expressed gene (DEG) in non-stimulated monocytes after 6mo of drinking. (B, C) Functional enrichment of DEG in non-stimulated monocytes at baseline (B) and after 6mo of heavy ethanol use (C). (D) Violin plot of selected DEG from non-stimulated monocytes. Genes below the * are significantly more expressed in the baseline samples, and genes above are significantly more expressed after 6mo of heavy drinking. (E) Volcano plot depicting DEG in LPS stimulated monocytes after 6mo of drinking. (F, G) Functional enrichment of DEG in LPS-stimulated monocytes at baseline (F) and after 6mo of heavy ethanol use (G). (H) Violin plot of selected DEG from LPS stimulated monocytes. Genes below the * are significantly more expressed in the baseline samples, and genes above are significantly more expressed after 6mo of heavy drinking. (I) Functional module scores between sampling time points at rest (left) and with LPS stimulation (right). Differentially expressed genes were determined via DESeq2 under default settings in Seurat and enriched using Metascape. Only statistically significant genes (average log(fold-change) cutoff >0.58 or <-0.58; adjusted p-value ≤ 0.05) were included in downstream analysis. Values for module scores were determined using an unpaired t-test with Welch’s correction. All bar plots depict the mean ± SEM. A p-value of <0.05 was considered significant.

Figure 5

Ethanol consumption and LPS stimulation drive monocyte differentiation along different trajectories. (A) UMAP indicating the two distinct lineage trajectories identified using Slingshot. (B, D) Progression graph showing the density of samples across pseudo time and the (C, E) differential gene expression (DEG) between conditions along its trajectory, respective to Lineage 1 (B, C) and Lineage 2 (D, E).

Figure 6

Heavy ethanol consumption for 6 months induces histone modification in circulating monocytes. (A) Bar plots representing the expression of histone modification determined via ELISA before and after 6mo of ethanol. (B) Number of differentially abundant peaks identified with H3k4me3 before and after 6mo of heavy drinking. (C) Genomic distribution of peaks. (D, E) Enrichment of peaks more abundant in the promoter region at baseline and (E) example pileups of respective genes. (F) Average expression of genes from scRNAseq. (G) Bar graph of TLR4 signaling pathway module score. (H) Enrichment of peaks more abundant in the promoter region at 6mo. Statistical significance was determined for histone modifications using a linear mixed model. Differential Peak analysis for CUT&Tag was completed using the HOMER’s getDifferentialPeaks function with a cutoff of Poisson-based p-value < 0.0001. Values for module scores were determined using an unpaired t-test with Welch’s correction. All bar plots depict the mean ± SEM. A p-value of <0.05 was considered significant.