Female and male Wild type (WT) C57BL/6 (stock # 000664) and Interferon α/β receptor 1-knockout (IFNR^KO) (stock # 32045-JAX) mice were purchased from Jackson Labs, Sacramento, CA. Six-week-old mice received ethanol (EtOH) (Pharmaco, Greenfield Global, Brookfield, CT) in water as per Meadows-Cook model (MC) and normal chow ad libitum for four weeks. EtOH concentration increased from 0% to 20% (v/v) gradually (5%, 10% and 15% for 4 days interval between each change then up to 20% in the 3^rd week). Mice were placed on MC diet for 4 weeks. Liver and body weight were measured, and whole blood, BM, liver, spleen, and lungs were collected for analysis. All mice were kept in an infection-free environment, and only control mice had access to alcohol-free water. All experiments were approved by the Rush University Medical Center, Institutional Animal Care & Use Committee and performed following the guidelines of the National Institutes of Health.

Whole blood was obtained by cardiac puncture. After clot formation, samples were centrifuged and sera were obtained and used for analysis. Hepatic leukocyte isolation was performed as described previously (22) with minor modifications. Briefly, to eliminate circulating leukocytes in livers, each mouse was perfused with 10 mL of cold PBS (Fisher bioreagents, Pittsburgh, PA) via the portal vein. Lungs and livers were harvested and cut into small pieces (about 1 mm²) on ice then pushed through a 70 µm filter (Biologix group limited, Shandong, China). After washing, pellets were resuspended in 8 ml of 40% Percoll (GE Healthcare, Waukesha, WI) and layered on top of 3 ml of 70% Percoll and centrifuged at 900 x g for 25 minutes at room temperature. Leukocytes were collected from the interface of 40% and 70% Percoll for counting and staining. Splenocytes were collected by mashing the spleen through a 70 µm filter. BM cells were flushed with 10 mL of cold PBS from the femur and tibia. Red cell lysis of spleen and BM cells were performed before counting and staining.

Samples were stained with fixable viability dye eFluor 506 eBioscience (Thermo Fisher Scientific, Waltham, MA) and fluorophore-conjugated antibodies against mouse CD16/32, CD45, CD45R/B220, CD3, CD19, NK1.1, Ly-6C, Ly-6G, PDCA-1, MHC-II, CD11b, CD11c, CD103, CD115, CD117, CD127, CD135, F4/80, Gr-1, Ki-67, Sca-1, and TER-119 ( Supplementary Table 1 ). Unless stated otherwise, all cell populations were gated as shown in Supplementary Table 2 . Ki-67 relative mean fluorescent intensity (MFI) was calculated by dividing the MFI of each sample by the average MFI of the female control. Samples were acquired using an LSRFortessa flow cytometer (BD Bioscience, San Jose, CA). FlowJo software version 10.0.8r1 (FlowJo, LLC. Becton, Dickinson and Company, Franklin Lakes, NJ) was used for the analysis of the acquired flow cytometry data.

Blood was obtained by cardiac puncture at the time of euthanasia and centrifuged at 6,000 x g for 5 min for serum collection. Serum macrophage-colony stimulating factor (M-CSF) was measured by ELISA (MMC00, R&D systems, Minneapolis, MN). Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were measured using the ALT and AST Reagents (7526 and 7561, Pointe Scientific, Canton, MI). Serum LPS (endotoxins) was measured, according to manufacturer instruction, by PYROGENT-5000 Kinetic Turbidimetric LAL Assay (N383, Lonza, Morristown, NJ)

Hepatic triglycerides were measured using the triglyceride assay kit according to the protocol provided by the manufacturer (ab65336, Abcam, Cambridge, MA).

Total hepatic mRNA was isolated from flash frozen liver in liquid nitrogen. The University of Illinois at Chicago (UIC) Genomics Research Core processed the specimens for total mRNA extraction, cDNA synthesis. qRT-PCR for cytokines, chemokines, and adhesion molecules implicated in monocyte trafficking was done, using primers described in Supplementary Table 3 , on the ViiA 7 Real-Time PCR System (Applied Biosystems, Foster City, CA). Fold change was calculated using the comparative Ct method. Housekeeping genes used are beta-Actin (Actb) and beta-2-microglobulin (B2m).

All figures generated and all statistical analyses were done using GraphPad Prism version 8.3.0 (GraphPad Software, Inc. San Diego, CA). Two way ANOVA with multiple comparisons is used to calculate the p-values. A p-value of ≤ 0.05 was considered significant. The level of significance indicated by asterisks as follow: *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Unless stated otherwise, the data are presented as means ± standard error of the mean (SEM).

To investigate the effect of alcohol consumption on monocytes homeostasis, female and male mice received alcohol in drinking water per Meadows-Cook (MC) model, a well-accepted mouse model of chronic alcohol consumption to characterize immunological effects of alcohol (23, 24). We chose MC, not Lieber DeCarli (LD), the most common mouse model of chronic alcoholic liver injury, because we are interested in studying the specific effect of alcohol on immune cells without the interference of the high fat and liquid diet used in LD model. Moreover, previous work published by our group demonstrated that LD model shared with alcohol exposure alone only a very restricted number of altered pathways (24).

Four weeks of alcohol consumption caused a significant reduction in the body weight, but not liver weight, of male mice; however, normalized liver/body weight ratio, surrogate for steatosis development, showed a significant increase only in female mice compared to controls ( Supplementary Figures 1A–C ). Despite no significant changes in liver triglycerides and liver enzymes, histological examination showed very mild small droplet steatosis in only alcohol-fed female mice compared to their controls ( Supplementary Figures 1D, E and Supplementary Figures 2A–C ). Serum levels of LPS in alcohol-fed female mice increased but did not reach statistical significance when compared with female control mice ( Supplementary Figure 2D ). Also, after alcohol consumption, the serum level of LPS in female mice is higher than in male mice but without statistical significance ( Supplementary Figure 2D ). Hepatic leukocytes did not increase in alcohol-fed mice in females and males ( Supplementary Figure 3 ). Interestingly, female mice have more hepatic leukocytes than males in only alcohol-fed mice ( Supplementary Figure 3 ). The presence of small droplet discrete liver steatosis and increased liver/body weight ratio support early histological evidence of alcoholic liver injury in female mice, while the increase in number of hepatic leukocytes in alcohol-fed female mice compared to males may represent a potential modulator for susceptibility to alcoholic liver injury in females (19–21). We have shown that alcohol has a sex specific effect on innate immune cells after 3 months of alcohol exposure (25) so we wanted to further study innate immune cellular compartments affected during this earlier time points.

Hepatic leukocytes were isolated and stained for flow cytometry analysis and gated to identify hepatic monocytes, neutrophils, and DCs ( Supplementary Figure 4 ). Females that consumed alcohol for four weeks showed a significantly greater increase in hepatic monocytes than controls and alcohol-fed male mice ( Figure 1A ). Since these hepatic monocytes were isolated from perfused livers we will refer to them from here on as monocyte derived macrophages (MDM). However, we observed no statistically significant change in hepatic neutrophils and DCs ( Supplementary Figure 5 ) between alcohol-fed and control mice in both female and male groups at this early time point. These data demonstrated that one of the earliest immunological innate events induced by alcohol in mice is identified in the liver at the level of monocytes compartment and present only in females. This sex dependent immunological effect of alcohol on liver monocytes combined with the well-known dichotomy in susceptibility of alcoholic immune effects triggered us to explore the mechanism of alcohol exposure on hepatic monocytes.

Classically in mice, monocytes are divided into two primary subsets based on the phenotypic expression of Ly-6C, CCR2, and CX3CR1 (1, 2). Monocytes expressing high levels of Ly-6C and CCR2 (Ly-6C^hiCCR2⁺) rapidly migrate to sites of inflammation to give rise to inflammatory macrophages and DCs and are known as “inflammatory monocytes” (1, 2). The locally patrolling monocytes express high levels of CX3CR1 but low levels of both Ly-6C and CCR2 and described phenotypically as Ly-6C^loCX3CR1⁺ (1, 2). Investigating the expression of Ly-6C on hepatic monocytes revealed a significant increase only in Ly-6C^hi and Ly-6C^lo MDMs in alcohol-fed female mice compared to controls ( Figures 1B, C ). Also, in alcohol consumed mice, females have more Ly-6C^hi and Ly-6C^lo MDMs than males ( Figures 1B, C ). Additionally, histological examination using immune-fluorescent microscopy showed a reduction in Ly-6C staining in the liver tissue from female mice consumed alcohol and controls compared to their male counterparts ( Figure 1D ). These data indicate that the hepatic MDMs increase is the result of the expansion of both Ly-6C^hi and Ly-6C^lo subset numbers.

Monocyte/Macrophage colony-stimulating factor (M-CSF), a growth factor shown to be essential for monocyte homeostasis and differentiation (2, 26). Mice deficient in CD115, the M-CSF receptor, have deficiencies in monocytes and macrophages (2, 26). Therefore, we tested the serum level of M-CSF. Alcohol consumption induced a reduction in the serum level of M-CSF only in female mice compared to controls ( Figure 2A ). Additionally, control female mice showed higher levels of serum M-CSF compared to their male counterparts ( Figure 2A ). Studying BM leukocytes did not show a significant increase in alcohol-fed mice compared to controls ( Supplementary Figure 6 ). Interestingly, there was a significant decrease in BM leukocytes in females compared to males in control groups ( Supplementary Figure 6 ). Considering the increase in hepatic leukocytes in females compared to males ( Supplementary Figure 3 ), these data might indicate an organ-specific nature of the potential immune-cellular mechanism(s) that might explain the higher susceptibility of females to develop alcoholic tissue injury. We studied the classically identified BM progenitors gated as shown in Supplementary Figures 7 and 8 . The data did not reveal a change in granulocyte-macrophage progenitor (GMP) ( Figure 2B ) and macrophage-dendritic cell progenitor (MDP) ( Figure 2C ) in alcohol-fed mice compared to controls. Investigating the expression of Ki-67, a cell proliferation marker (27), in GMP and MDP did not reveal significant changes after alcohol consumption ( Figures 2D, E ). These data suggest that the increase in monocytes found in female mice is not a result of alcohol interference in the development of BM classical progenitors.

We asked whether the increase in hepatic MDMs results in a change of hepatic cytokine expression and whether this reflects a change in the hepatic chemokines and adhesion molecules gradient implicated in the recruitment and migration of monocytes (28). The cytokine tumor necrosis factor alpha (TNFα) was upregulated in the livers of female mice fed alcohol compared to their control counterparts ( Figure 3A ). The expression of chemokines, chemokine receptors, and adhesion molecules tested did not reveal significant changes between alcohol-fed and control female mice ( Figures 3A–C ). The increase in TNFα data is consistent with overwhelming evidence in the literature about the implications of TNFα in alcohol-induced liver injury in both humans and mice and marks an early event in alcohol immunopathology (29–33).

Our data did not reveal a disturbance in BM progenitors or recruitment and migration molecules that might explain the increase in MDMs we found in female mice that consumed alcohol. Published data showed regulation of emergent monocytopoiesis by type I interferon (IFN-I) signaling (34). Therefore, we used Interferon α/β receptor 1-deficient (IFNR^KO) mice to study the effect of alcohol on hepatic monocytes in the absence of IFN-I signaling. Interestingly, IFNR^KO female mice consumed alcohol for four weeks failed to show an increase in their hepatic MDMs compared to control-fed mice, as did their WT counterparts ( Figures 1A–C ). In the livers of females, there are more Ly-6C^hi and Ly-6C^lo MDMs in WT compared to IFNR^KO, in alcohol-fed mice ( Figures 1B, C ). This data suggest a critical role for IFN-I signaling in the increase of MDMs induced by alcohol in female mice.

We further investigated the BM development of monocytes in both WT and IFNR^KO mice. First, we investigated mature monocytes in the BM, and in both WT and IFNR^KO mice, our data suggest no change in monocyte subsets between alcohol-fed mice and controls ( Figures 4A–C ). Comparing BM progenitors between alcohol-fed and control mice did not show significant changes in both WT and IFNR^KO mice ( Figures 4D–F ). However, comparing the same progenitors between WT and IFNR^KO mice revealed a significant reduction in MDP in IFNR^KO female control mice ( Figure 4E ). In IFNR^KO, both control and alcohol-fed female mice, compared to WT, there is a reduction in common monocyte progenitors (cMoP), but it did not reach statistical significance ( Figure 4F ). These data further demonstrate a potential role of IFN-I signaling in the increase of Ly-6C^hi and Ly-6C^lo subsets in female mice early during alcohol consumption and preceding histological findings of alcoholic liver injury.