Regular Lieber-DeCarli diet and maltodextrin were obtained from Dyets Inc (Cat# 710260; Bethlehem, PA, USA) and Bioserv (Flemington, NJ, USA), respectively. Corticosterone was procured from Sigma-Aldrich (St Louis, MO, USA). Antibodies for occludin and ZO-1 were procured from Thermo Fisher Scientific (Waltham, MA, USA). Antibodies against E-cadherin and β-catenin were procured from BD Biosciences (San Jose, CA, USA). Secondary antibodies conjugated with AlexaFluor-488 or Cy3 were procured from Molecular Probes (Eugene, OR, USA). All other antibodies or chemicals were procured from either Thermo Fisher Scientific or Sigma-Aldrich (St Louis, MO, USA).

Mouse strain deficient in the TRPV6 gene was a generous gift from Prof. Matthias A. Hediger, University of Bern, Switzerland), and acquired through Dr. Sylvia Christakos (Rutgers New Jersey Medical School, Newark, NJ, USA). TRPV6 mice have been backcrossed to C57BL/6 mice for more than 20 generations. The wild-type mice were derived from the same colony and are littermates. The Institutional Animal Care and Use Committee at UTHSC have approved all protocols and guidelines for animal experiments described in this study. Animals (two per cage) were housed in a 12:12 h light: dark cycle conditions in our institutional core facility. Animals had unrestricted access to standard laboratory chow and water before the experiments.

Chronic EtOH feeding: Adult Trpv6^-/- and its corresponding wild-type mice (age: 8–10 weeks, male and female) were fed Lieber–DeCarli liquid diet containing varying concentrations of EtOH (0% for two days, 1% for two days, 2% for two days, 4% for one week, 5% for one week, and 6% for one week) for four weeks as described previously (21, 40). In control groups, EtOH was substituted with isocaloric maltodextrin. Diet consumption was recorded daily, and body weight measurements were documented twice a week. The pair-fed and EtOH-fed mice were subjected to one of the two kinds of stressors.

The mucosal permeability of the colon and ileum was assessed as described previously (20). At the end of the study, mice were restrained and injected with FITC-inulin (100 mg/ml solution: 2 μl/g body weight) through their tail veins. One hour following infusion, blood samples were drawn by cardiac puncture under isoflurane anesthesia to prepare plasma, followed by euthanizing mice by cervical dislocation. The 0.9% saline solution was used for flushing the intestinal segments’ luminal contents. Fluorescence in plasma and luminal flushing samples was measured using a fluorescence plate reader. Fluorescence in the luminal flushing was correlated with the plasma fluorescence; results are presented as percentages of the inulin load to mice. There are multiple ways to measure intestinal permeability in mice in vivo. Each of them has advantages and disadvantages. Luminal-to-vascular and vascular-to-luminal permeability measurements show no differences as permeability through leaky tight junction has no directionality. The vascular-to-luminal permeability measurements involve the minor manipulation of the intestine, and we have consistently shown an excellent correlation between permeability measured by this method and TJ disruption by confocal microscopy in corresponding mice in previous studies.

Colon cryosections (10 μm) were fixed in an acetone and EtOH (1:1) mixture at −20°C for 2 minutes and then rehydrated in phosphate-buffered saline (PBS). Fifteen minutes of permeabilization with 0.2% Triton X-100 in PBS was followed by blocking with 4% nonfat milk in Triton-Tris buffer. Next, a primary antibody specific for TJ and AJ (1:100 dilution) was added and incubated for one hour, followed by incubation with secondary antibodies for an additional hour. After incubations with antibodies, the section was incubated with Hoechst 33342 for 10 minutes, and fluorescence images were captured using a Zeiss 710 confocal microscope (Carl Zeiss Microscopy, Jena, Germany). Images were stacked and processed using Image J software (NIH, Bethesda, MD) and Adobe Photoshop (Adobe Systems Inc., San Jose, CA).

The analysis of particular mRNA was performed by RT-PCR as described previously (37). cDNAs were synthesized from 1.5 μg of RNA using the ThermoScript RT-PCR instrument for first-strand synthesis (Invitrogen). Quantitative PCR (qPCR) reactions were performed using a cDNA mix with primers in a final volume of 25 μl in an Applied Biosystems Quant Studio 6-Flex Real-Time PCR instrument (Norwalk, CT, USA). The parameter of the cycle was as follows: 50°C for two minutes, one step of denaturation at 95°C for ten minutes, and 40 cycles of denaturation at 95°C for ten seconds, followed by annealing and elongation at 60°C. The ΔΔCt analysis normalized each transcript’s relative expression level to that of GAPDH. Table S1 of the Supplemental Information presents the primer sequences used for qPCR.

The microbiome of colonic flushing was analyzed using 16S rRNA sequencing and metagenomic analysis, as described in the previous study (20).

Using a Pierce LAL Chromogenic Endotoxin Quantitation Kit (Thermo Scientific), plasma endotoxin concentration was measured as described in our previous studies (20, 37).

Plasma cytokine and chemokine levels were evaluated using a Duoset ELISA kit (R&D system, Minneapolis, MN) according to the manufacturer’s instructions. Briefly, 50 μL of plasma was incubated overnight on microplates treated with capture antibody, followed by a 2-hour incubation with detection antibody. The sample was then probed for 20 minutes with streptavidin conjugated with horseradish peroxidase and substrate solution. After terminating the reaction with a stop-solution, the absorbance was determined at 450 nm and normalized for wavelength at 570 nm.

Liver tissues were fixed in 10% buffered formalin, and 8 μm-thick paraffin-embedded slices were stained with hematoxylin and eosin (H&E) as detailed previously (20, 37). Stained sections were imaged using a Nikon 80Ti microscope (Nikon Instruments, Inc., Melville, NY) with a 10x objective lens and a color camera.

The activities of plasma aspartate transaminase (AST) and alanine transaminase (ALT) were determined using a colorimetric assay using ALT (Cayman, Cat#700260) and AST (Cayman, Cat#7012640) assay kits, respectively, according to the vendor’s instructions.

The triglyceride content in the liver was assessed by the GPO method using the previously described assay kit from Pointe Scientific Inc. (Canton, MI) (20, 37). Briefly, Tissue lipids were extracted by treating the sample in 3 M potassium hydroxide (in 65% EtOH) at 70°C for one hour and at room temperature overnight. Enzymatic hydrolysis of triglycerides into glycerol and free fatty acids was followed by spectrophotometric determination of glycerol at 540 nm. The hepatic triglyceride values were represented as mg triglyceride/g liver tissue.

All data are presented as Mean ± SEM. ANOVA was used to examine the differences between several groups (Prism 6.0). Tukey’s t-test was employed to determine the statistical significance between different testing sets and the respective control. The statistical significance was determined with 95% confidence.

The values are shown as the Mean ± standard error of the mean (SEM) (n =4-6); * p <0.05, ** p <0.01, *** p <0.005, and **** p <0.001 signify a significant difference between the groups that are specified; ns denotes that the difference is not significant.

Corticosterone disrupts the intestinal mucosal barrier and potentiates EtOH-induced barrier dysfunction (20). EtOH and AA disrupt TJ in Caco-2 cell monolayers by an intracellular calcium-dependent mechanism (15). TRPV6 is essential for alcohol-induced calcium influx in the intestinal mucosa, TJ breakdown, endotoxemia, systemic inflammation, and liver damage (18). To determine whether TRPV6 plays a role in corticosterone-induced intestinal mucosal injury and corticosterone-induced potentiation of alcohol effects on the gut and liver, we fed wild-type and Trpv6^-/- mice EtOH for four weeks. In wild-type mice, EtOH-feeding enhanced mucosal permeability to FITC-inulin (6 kDa) in the colon ( Figure 1A ) and ileum ( Figure 1B ). Corticosterone, by itself, also significantly elevated the colonic mucosal permeability, and it synergistically potentiated the EtOH-induced increase in colonic mucosal permeability. Trpv6^-/- mice resisted corticosterone-induced intestinal mucosal permeability and corticosterone-mediated promotion of alcohol-induced mucosal permeability. An increase in mucosal permeability suggested that corticosterone may disrupt TJ and AJ. Therefore, cryosections of the distal colon were stained for TJ and AJ proteins. The presence of occludin and ZO-1 at the intercellular junctions of the intestinal epithelium of pair-fed wild-type and Trpv6^-/- mice indicates the high integrity of TJ in pair-fed animals. Corticosterone alone elicited a modest displacement of occludin and ZO-1 from epithelium in wild-type mice. The TJ disruption was more pronounced when EtOH feeding was combined with corticosterone administration than EtOH or corticosterone alone ( Figure 1C ). Densitometry of fluorescence at the epithelial junctions confirmed a significant decrease in the junctional distribution of ZO-1 in EtOH-fed mice, which was more severe when combined with corticosterone treatment ( Supplementary Figure S2A ). Corticosterone also reduced E-cadherin and β-catenin fluorescence in the colonic epithelium of wild-type mice but not in Trpv6^-/- mice ( Figure 1D ); confirmed by densitometry of β-catenin at the epithelial junctions ( Supplementary Figure S2B ). The impact of EtOH and corticosterone on the redistribution of E-cadherin and β-catenin appears to be more severe in wild-type mice; such an effect of EtOH and corticosterone was absent in the Trpv6^-/- mice.

We measured IL-6 and MCP1 mRNA in the colonic mucosa as an indicator of mucosal inflammatory response. EtOH significantly increased IL-6 ( Figure 1E ) and MCP1 ( Figure 1F ) mRNA. Corticosterone increased MCP1 mRNA and potentiated EtOH-induced increases in MCP1 mRNA in the wild-type mice but not in Trpv6^-/- mice ( Figure 1E ). Corticosterone reduced IL-6 mRNA, but it synergistically potentiated the EtOH-induced increase in IL-6 mRNA ( Fig 1F ). The effects of corticosterone and EtOH-induced colonic inflammatory response were absent in Trpv6^-/- mice. These data suggest that corticosterone exacerbates the alcohol-induced intestinal epithelial barrier dysfunction and mucosal inflammatory response by a TRPV6-dependent mechanism.

The diet intake was similar in all groups except the corticosterone treatment groups ( Supplementary Figure S1A ). Similar diet intake in EtOH-fed groups (with or without corticosterone) in wild-type and Trpv6^-/- mice indicate that the EtOH consumption was similar in these strains. Neither EtOH nor corticosterone treatment altered the body weight changes during the experiment ( Supplementary Figure S1B-E ); corticosterone treatment alone tends to increase the body weight; however, the differences were not statistically significant. In addition, the colon length was considerably shortened by EtOH feeding in wild-type mice, whereas it was slightly increased when EtOH was combined with corticosterone treatment ( Supplementary Figure S1F ); such differences in colon length were absent in Trpv6^-/- mice.

Epithelial TJ disruption allows LPS absorption from the intestinal lumen into the portal and systemic circulation, leading to elevated plasma LPS, a condition referred to as endotoxemia (6). Therefore, we assessed the plasma LPS as an indicator of endotoxemia. Results show that EtOH and corticosterone significantly increased plasma LPS in wild-type mice ( Figure 2A ). The combined EtOH and corticosterone treatments induced a synergistic elevation of plasma LPS, the effect more remarkable than the sum of EtOH and corticosterone-alone effects. This effect of corticosterone and corticosterone-mediated potentiation of EtOH effects on plasma LPS were absent in Trpv6^-/- mice. LPS in the circulation stimulates immune cells in multiple organs resulting in a systemic inflammatory response. Therefore, we evaluated the plasma concentration of representative cytokines as the indicator of systemic inflammation. EtOH or corticosterone alone significantly elevated plasma IL-6 in the wild-type mice. Plasma IL-6 in pair-fed Trpv6^-/- mice were slightly, but significantly, more significant than that in pair-fed wild-type mice; however, neither EtOH nor corticosterone elevated plasma IL-6 in Trpv6^-/- mice ( Figure 2B ). Corticosterone treatment elevated plasma MCP1, whereas EtOH feeding caused a slight decrease ( Figure 2C ). However, EtOH and corticosterone together produced a synergistic elevation of plasma MCP1. Plasma MCP1 in pair-fed Trpv6^-/- mice was considerably more significant than in pair-fed wild-type mice. However, EtOH, corticosterone, or EtOH + corticosterone failed to increase plasma MCP1 levels ( Figure 2C ). These findings suggest that TRPV6 is required for corticosterone- and corticosterone-mediated potentiation of EtOH-induced endotoxemia and systemic inflammatory response.

Endotoxemia and intestinal barrier disruption play essential roles in the pathophysiology of alcohol-related liver damage. The liver is the initial organ that receives portal LPS. Therefore, we assessed liver damage in wild-type and Trpv6^-/- mice. In wild-type mice, the plasma levels of liver injury markers ALT ( Figure 3A ) and AST ( Figure 3B ) were elevated by EtOH and corticosterone. EtOH and corticosterone in combination caused synergistically higher levels of plasma ALT and AST in wild-type mice. These effects of EtOH and corticosterone (alone or together) on plasma ALT and AST were minimal in Trpv6^-/- mice. The changes in the plasma liver markers were associated with parallel changes in liver histopathology. In wild-type mice, EtOH feeding revealed significant liver histopathological changes characterized by white droplets of varying sizes, representing the fat deposits predominantly in the centrilobular zone, a condition referred to as fatty liver ( Figure 3C ). In wild-type mice, corticosterone alone did not cause gross histopathological changes, but it increased the severity of EtOH-induced histopathological changes in wild-type mice. These effects of EtOH were absent in Trpv6^-/- mice in the absence or presence of corticosterone. EtOH-induced fatty liver was confirmed by measuring the liver triglyceride content. In wild-type mice, EtOH and corticosterone alone significantly increased liver triglyceride; however, the increase was much more significant when EtOH feeding was paired with corticosterone treatment ( Figure 3D ). These effects were minimal and non-significant in Trpv6^-/- mice. These findings suggest that TRPV6 is necessary for EtOH- and corticosterone-mediated potentiation of EtOH-induced liver injury. EtOH-feeding significantly reduced liver weights relative to body weights in the absence of corticosterone in wild-type mice ( Supplementary Figure S1G ); however, liver weight was increased by EtOH in the presence of corticosterone. These changes in liver weights were absent in Trpv6^-/- mice.

To investigate endogenous corticosterone’s effect, we examined chronic restraint stress (CRS) on gut and liver injury and potentiation of alcohol-induced gut mucosal barrier dysfunction, endotoxemia, and liver injury in wild-type and Trpv6^-/- mice. In wild-type mice, EtOH feeding significantly elevated mucosal inulin permeability in the colon ( Figure 4A ) and ileum ( Figure 4B ). CRS also elevated the mucosal permeability to inulin in the ileum and colon. The combination of EtOH feeding and CRS resulted in a much more severe increase in mucosal permeability in wild-type mice; the effect was more significant than the sum of the effects of EtOH and CRS by themselves ( Figures 4A, B ). The effects of EtOH and CRS, individually or combined, on mucosal permeability were absent in Trpv6^-/- mice. Increased mucosal permeability suggested that EtOH and CRS disrupt TJ and AJ. Results indicate that TJ proteins (occludin and ZO-1) are co-localized at the intercellular junction of colonic epithelium in both pair-fed wild-type and Trpv6^-/- mice. In the wild-type mice, EtOH or CRS alone induced a moderate junctional redistribution of TJ from the epithelial junctions; these effects were much more severe in the combined EtOH and CRS treatments compared to EtOH or CRS group ( Figure 4C ). Densitometry of fluorescence at the epithelial junctions confirmed a significant decrease in the junctional distribution of ZO-1 in EtOH-fed mice ( Supplementary Figure S4A ). EtOH and CRS also reduced AJ proteins (E-cadherin and β-catenin) fluorescence in the colonic epithelium of wild-type mice but not in Trpv6^-/- mice ( Figure 4D ). The effects of EtOH and CRS combination on the E-Cadherin and β-catenin distribution appears to be more severe in wild-type mice, and such an effect of EtOH and CRS was absent in Trpv6^-/- mice; confirmed by densitometry of β-catenin at the epithelial junctions ( Supplementary Figure S4B ).

We measured IL-6, TNFα, and MCP1 mRNA in the colonic mucosa as the indicators of mucosal inflammatory response. EtOH increased IL-6 ( Figure 4E ), TNFα ( Figure 4F ), and MCP1 mRNA ( Figure 4G ). CRS also increased IL-6, TNFα, and MCP1 mRNA and potentiated EtOH-induced increases in cytokine mRNA expression in wild-type mice but not in Trpv6^-/- mice. These data suggest that CRS exacerbates the alcohol-induced intestinal epithelial barrier dysfunction and mucosal inflammatory response by a TRPV6-dependent mechanism. The diet intake was similar in all groups, except for a slightly higher diet intake in EtOH-fed Trpv6^-/- mice compared to that in EtOH-fed wild-type mice ( Supplementary Figure S3A ). EtOH-feeding significantly reduced body weight gain in wild-type mice during treatments; this effect was independent of CRS treatment ( Supplementary Figure S3B–E ). EtOH-induced weight loss was absent in Trpv6^-/- mice. EtOH-supplementation dramatically increased the colon length in wild-type mice, irrespective of CRS ( Supplementary Figure S3F ); EtOH-induced colon length increases were absent in Trpv6^-/- mice.

Disrupted epithelial TJ permits LPS to enter the mucosa from the intestinal lumen, causing endotoxemia and triggering a systemic inflammatory response. Results show that EtOH and CRS alone significantly increased plasma LPS in wild-type mice ( Figure 5A ). The combined EtOH-feeding and CRS induced a synergistic elevation of plasma LPS, the effect more significant than the sum of EtOH and CRS effects. These effects of CRS and CRS-mediated potentiation of EtOH-induced elevation of plasma LPS were absent in Trpv6^-/- mice. EtOH and CRS alone significantly elevated plasma IL-6 ( Figure 5B ) and MCP1 ( Figure 5C ) in the wild-type mice; the cytokine levels were significantly higher when EtOH and CRS were combined. The effects of EtOH and CRS on plasma cytokines, alone or in combination, were minimal in Trpv6^-/- mice. These data suggest that TRPV6 is required for CRS and CRS-mediated potentiation of EtOH-induced endotoxemia and systemic inflammation.

The prior investigation established that CRS exacerbated EtOH-induced liver damage in murine models (37). Here, we confirm the CRS effect on liver damage in wild-type mice. The plasma ALT ( Figure 6A ) and AST ( Figure 6B ) activities were elevated by EtOH-feeding and CRS alone in wild-type mice. EtOH and CRS, in combination, caused synergistically higher levels of plasma ALT and AST in wild-type mice. No significant effects of EtOH and CRS (alone or together) on plasma ALT and AST were recorded in Trpv6^-/- mice. Histopathological analyses of the EtOH-fed wild-type liver showed significant pathologic changes characterized by varying-sized white droplets, representing the fat deposits predominantly in the centrilobular zone ( Figure 6C ). CRS alone did not cause gross histopathological changes, but it increased the severity of EtOH-induced histopathologic changes in wild-type mice. These effects of EtOH alone or combined with CRS were absent in Trpv6^-/- mice. EtOH-induced fatty liver was confirmed by measuring the liver triglyceride content. In wild-type mice, EtOH and CRS alone significantly increased liver triglyceride; however, the increase was significantly greater when EtOH supplementation was combined with CRS ( Figure 6D ). These effects were minimal and non-significant in Trpv6^-/- mice. These findings suggest that TRPV6 is necessary for EtOH- and CRS-mediated potentiation of EtOH-induced liver injury. EtOH-feeding significantly increased liver weights relative to body weights in wild-type mice, irrespective of CRS ( Supplementary Figure S3G ); EtOH-induced liver weight increases were absent in Trpv6^-/- mice.

We next analyzed the fecal microbiota to assess the effects of EtOH-feeding and CRS gut microbiome composition and the role of TRPV6 in these effects. At the phylum level, samples from all groups contained a high relative abundance of Bacteroidetes and Firmicutes, followed by lower levels of Proteobacteria ( Figure 7A ). As assessed by Shannon Index, the alpha diversity was reduced by EtOH with or without combination with CRS; CRS alone did not cause a significant change in alpha diversity ( Figure 7B ). In contrast, no alpha diversity changes were observed with any treatments in Trpv6^-/- animals. Assessment using Bray-Curtis Index displayed by Principal Coordinate Analysis (PCoA) ordination plots demonstrated no significant effects on beta diversity within genotypes ( Figure 7C ). However, a supervised redundancy analysis (RDA) on all groups showed separation of wild-type EtOH and EtOH + CRS groups compared with the other groups, which clustered together except for pair-fed Trpv6^-/- mice ( Figure 7D ). To better assess the specific changes driving decreased alpha diversity in wild-type EtOH alone and in combination with CRS, we performed genus-level analysis using Spearman rank correlation coefficients with heatmaps, pattern search analysis, and network analysis in wild-type and Trpv6^-/- mice. The Spearman rank correlation analysis demonstrated that wild-type EtOH-fed with or without CRS contained an elevated abundance of Bilophila, Staphylococcus, and Lactobacillus ( Figure 7E ). Specifically, wild-type EtOH + CRS treated mice also contained a markedly higher relative abundance of Odoribacter, Rhodococcus, Dehalobacterium, Ruminococcus, and unclassified Peptococcaceae, Helicobacteraceae, and Clostridiales. These patterns were supported by the results of pattern search analysis ( Supplementary Figure S5A ) and network analysis ( Supplementary Figure S4A ) in wild-type mice. Similar taxa patterns were not observed in Trpv6^-/- mice under the same treatment conditions ( Figure 7E , Figure S5B , Figure S6B ). Together these results suggest unique enrichment of microbes associated with pro-inflammatory and barrier dysfunction conditions in EtOH-fed with or without CRS in wild-type mice; such EtOH effects were absent in Trpv6^-/- mice. These data suggest that the TRPV6 channel is involved in the EtOH-induced alteration of gut microbiota.

In a previous study, we measured plasma corticosterone levels in CRS-induced and EtOH-fed mice. Data indicated that CRS and EtOH increased plasma corticosterone; however, the level was significantly higher when CRS and EtOH feeding were combined. Therefore, in response to the reviewer’s question, we have performed a corticosterone assay in plasma in the current studies to determine the corticosterone levels in TRPV6 knockout mice. Data show similar results in both Wild-type and Trpv6^-/- mice, indicating that the role of TRPV6 is downstream in glucocorticoid response ( Supplementary Figure S7 ) and not involved in corticosterone release.