Farnesoid X receptor enhances epithelial ACE2 expression and inhibits virally induced IL-6 secretion: implications for intestinal symptoms of SARS-CoV-2

Abstract

Intestinal inflammation and diarrhea are often associated with SARS-CoV-2 infection. The angiotensin converting enzyme 2 (ACE2) receptor plays a key role in SARS-CoV-2 pathogenesis, facilitating entry of the virus into epithelial cells, while also regulating mucosal inflammatory responses. Here, we investigated roles for the nuclear bile acid receptor farnesoid X receptor (FXR) in regulating ACE2 expression and virally mediated inflammatory responses in intestinal epithelia. Human colonic or ileal enteroids and cultured T₈₄ and Caco-2 monolayers were treated with the FXR agonists, obeticholic acid (OCA) or GW4064, or infected with live SARS-CoV-2 (2019-nCoV/USA_WA1/2020). Changes in mRNA, protein, or secreted cytokines were measured by qPCR, Western blotting, and ELISA. Treatment of undifferentiated colonic or ileal enteroids with OCA increased ACE2 mRNA by 2.1 ± 0.4-fold (n = 3; P = 0.08) and 2.3 ± 0.2-fold (n = 3; P < 0.05), respectively. In contrast, ACE2 expression in differentiated enteroids was not significantly altered. FXR activation in cultured epithelial monolayers also upregulated ACE2 mRNA, accompanied by increases in ACE2 expression and secretion. Further experiments revealed FXR activation to inhibit IL-6 release from both Caco-2 cells infected with SARS-CoV-2 and T₈₄ cells treated with the viral mimic, polyinosinic:polycytidylic acid, by 46 ± 12% (n = 3, P < 0.05) and 35 ± 6% (n = 8; P < 0.01), respectively. By virtue of its ability to modulate epithelial ACE2 expression and inhibit virus-mediated proinflammatory cytokine release, FXR represents a promising target for the development of new approaches to prevent intestinal manifestations of SARS-CoV-2.NEW & NOTEWORTHY Activation of the nuclear bile acid receptor, farnesoid X receptor (FXR), specifically upregulates ACE2 expression in undifferentiated colonic epithelial cells and inhibits virus-induced proinflammatory cytokine release. By virtue of these actions FXR represents a promising target for the development of new approaches to prevent intestinal manifestations of SARS-CoV-2 infection.

Keywords: ACE2; SARS-CoV-2; bile acid; farnesoid X receptor; intestinal epithelium.

Graphical abstract

Figure 1.

FXR activation upregulates ACE2 expression in undifferentiated colonic and ileal enteroids. Enteroids prepared from crypts isolated from human colonic and ileal tissue specimens were cultured as monolayers on semipermeable supports. Enteroids were studied in the undifferentiated state by exposure to medium containing Wnt, R-spondin, and noggin, or in the differentiated state by withdrawal of these growth factors for 5 days, and treated with OCA (10 µM) for 24 h. A: basal ACE2 expression in undifferentiated and differentiated colonic and ileal enteroids (n = 3; ***P < 0.001). Changes in ACE2 expression in response to OCA treatment in undifferentiated (B) and differentiated (C) colonic enteroids and undifferentiated (D) and differentiated (E) ileal enteroids were analyzed by bulk RNA-Seq, with results normalized to changes in the untreated conditions (n = 3; *P < 0.05). Results are expressed as means ± SE. Statistical analysis was performed using two-way ANOVA with Tukey’s multiple comparisons test or paired t test. FXR, farnesoid X receptor; OCA, obeticholic acid.

Figure 2.

FXR activation upregulates ACE2 expression and secretion in T₈₄ colonic epithelial cells. Serum-starved monolayers of T₈₄ colonic epithelial cells bilaterally treated with GW4064 (A) (5 µM) for 3–48 h (n = 8; *P < 0.05, **P < 0.01) or CDCA (B) (10 and 100 µM) for 48 h (n = 4; *P < 0.05) were analyzed for ACE2 mRNA expression by qRT-PCR. T₈₄ cells were treated with GW4064 (5 µM) at various time points between 3 and 72 h and aliquots of medium collected from the apical compartment of matched control and GW4064-treated cells were analyzed for levels of soluble ACE2 by ELISA (C) (n = 9; ***P < 0.001). D: ACE2 expression in the protein lysates was measured by Western blotting (n = 6; **P < 0.01). Results are expressed as means ± SE. Statistical analysis was performed using one-way ANOVA with Dunnett’s multiple comparisons test, two-way ANOVA with Šidák’s multiple comparisons test, or paired t test. FXR, farnesoid X receptor.

Figure 3.

FXR activation inhibits Poly I:C-induced IL-6 secretion and NF-κB signaling. T₈₄ cell monolayers were serum starved for 24 h and treated apically and basolaterally with either GW4064 (5 µM) or poly I:C (25 µg/mL) alone, or GW4064 in combination with poly I:C for 24 h. IL-6 secretion into the apical (A) and basolateral (B) compartments was measured by ELISA (n = 8; *P < 0.001, **P < 0.01, ***P < 0.001). HEK293 cells were transfected with constructs containing NF-κB response elements linked to the firefly luciferase reporter gene, FXR, RXR, or empty vector. A pRL-SV40 vector, containing Renilla luciferase, was used to control for transfection efficiency. Cells transfected with FXR/RXR (C) or with empty vector (D) were treated with either poly I:C (25 µg/mL) or poly I:C in combination with GW4064 for 24 h, after which they were lysed and NF-κB luciferase activity measured (n = 3; *P < 0.05). Results are expressed as means ± SE. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test. FXR, farnesoid X receptor.

Figure 4.

FXR activation upregulates ACE2 expression and inhibits SARS-CoV-2-induced IL-6 secretion from Caco-2 intestinal epithelial cells. Polarized monolayers of Caco-2 cells apically infected with SARS-CoV-2 were treated bilaterally with GW4064 (5 µM) for 120 h. Levels of ACE2 (A) and IL-6 (B) mRNA expression were measured by qPCR (n = 3; *P < 0.05, ***P < 0.001, ****P < 0.0001). Results are expressed as means ± SE. Statistical analysis was performed using two-way ANOVA with Tukey’s multiple comparisons test.