Vascular microRNA-204 is remotely governed by the microbiome and impairs endothelium-dependent vasorelaxation by downregulating Sirtuin1

Abstract

Gut microbiota promotes atherosclerosis, and vascular endothelial dysfunction, signalled by impaired endothelium-dependent vasorelaxation, is an early marker of atherosclerosis. Here we show that vascular microRNA-204 (miR-204) expression is remotely regulated by the microbiome, and impairs endothelial function by targeting the Sirtuin1 lysine deacetylase (Sirt1). MiR-204 is downregulated, while Sirt1 is upregulated, in aortas of germ-free mice. Suppression of gut microbiome with broad-spectrum antibiotics decreases miR-204, increases Sirt1 and bioavailable vascular nitric oxide, and improves endothelium-dependent vasorelaxation in mouse aortas. Antibiotics curtail aortic miR-204 upregulation, and rescue decline of aortic Sirt1 and endothelium-dependent vasorelaxation, triggered by high-fat diet feeding. Improvement of endothelium-dependent vasorelaxation by antibiotics is lost in mice lacking endothelial Sirt1. Systemic antagonism of miR-204 rescues impaired endothelium-dependent vasorelaxation and vascular Sirt1, and decreases vascular inflammation induced by high-fat diet. These findings reveal a gut microbe-vascular microRNA-Sirtuin1 nexus that leads to endothelial dysfunction.

Figure 1. MiR-204 is downregulated and Sirt1 is upregulated in aortas of GFM.

(a) Volcano plot of differentially expressed microRNAs in aortas of germ-free mice (GFM). Difference in the expression of aortic microRNAs between GFM and pathogen-free mice (PFM) is plotted on the x axis (log₂ scale), and significance is plotted on the y axis (log₁₀ scale). Upregulated or downregulated microRNAs are indicated in red and blue, respectively. n=3 mice for each microRNA. (b) Decreased mature and precursor miR-204 expression in aortas of GFM by qPCR (shown as % of PFM). n=4 for each group, *P<0.05 versus PFM. (c) In silico analysis of mRNAs targeted by microRNAs that were downregulated by nCounter assay in aortas of GFM. Number of downregulated microRNAs targeting a specific mRNA is shown on the y axis and % of total mRNA is shown on the x axis. MicroRNA-SVR score is colour-coded with threshold<−0.50 in grey, <−0.75 in blue and<−0.90 in red. Lower SVR score indicates stronger predicted microRNA–mRNA interaction. Sirt1 is targeted by microRNAs downregulated in GFM aortas, and based on the number of microRNAs targeting a specific gene it was predicted to be in top 0.5% of putatively regulated genes. (d) MiR-204 mimic (miR-204-M) decreases Sirt1 3′-UTR luciferase reporter activity. SC, scrambled control microRNA. *P<0.05 versus SC, n=5 independent experiments. (e) MiR-204-M (20 nM) decreases Sirt1 expression in human umbilical vein endothelial cells (HUVECs) in a time-dependent manner. Representative of three independent experiments is shown. (f,g) Increased Sirt1 (f) and decreased Trpm3 (g) mRNA expression in aortas of GFM (shown as % of PFM). n=4 for each group, *P<0.05 versus PFM. Independent sample t-test was used. Data shown as mean and error bar represents s.e.m.

Figure 2. Antibiotics improve vasorelaxation via endothelial Sirt1.

(a–c) Gut microbiota suppression in mice with broad-spectrum antibiotics downregulates aortic miR-204 (a: qPCR; b: in situ hybridization at × 100 magnification with miR-204 shown in purple with red nuclear counterstain, Scale bar; 20 μm) and Trpm3 (c). All data are shown as % of NPD. n=6 for NPD and NPD-AB, n=4 for NPD-AB-R. *P<0.05 versus NPD, ^#P<0.05 versus NPD-AB. (d,e) Gut microbiome suppression in mice with broad-spectrum antibiotics upregulates aortic Sirt1 and eNOS expression. All data are shown as % of NPD. n=3 for NPD and NPD-AB, n=4 for NPD-AB-R. *P<0.05 versus NPD, ^#P<0.05 versus NPD-AB. (f) Immunofluorescence of mouse aortic sections showing upregulation of total aortic and endothelial Sirt1 expression with antibiotics (magnification × 63, Scale bar; 20 μm). (g) Quantification of total aortic (n=4 sections) and endothelial (n=15 endothelial cells) Sirt1 immunostaining in f. (h,i) Gut microbiota suppression in mice with antibiotics improves endothelium-dependent vasorelaxation in wild-type C57Bl/6 (h), VE-Cad-Cre (i) and Sirt1^fl/fl (j) mice. NPD: n=10, NPD-AB: n=10, NPD-AB-R: n=8. *P<0.05 versus NPD, ^#P<0.05 versus NPD-AB. Sirt1^fl/fl: n=10, Sirt1^fl/fl-AB: n=20; *P<0.05 versus Sirt1^fl/fl. VE-Cad-Cre: n=5, VE-Cad-Cre-AB: n=9, *P<0.05 versus VE-Cad-Cre. (k) Gut microbiome suppression with antibiotics does not improve aortic endothelium-dependent vasorelaxation in eSirt^−/− mice. eSirt1^−/−: n=14, eSirt1^−/−AB: n=15. ^#P<0.05 versus eSirt1^−/−. (l) Adenovirus-mediated overexpression of Sirt1 lacking 3′-UTR rescues miR-204 mimic-induced impairment of endothelium-dependent vasorelaxation ex vivo. SC-AdLacZ: n=5; miR-204-AdLacZ: n=6; miR-204-AdSirt1: n=6. AdLacZ, control adenovirus-expressing E. coli LacZ; AdSirt1, adenovirus-expressing Sirt1; SC, scrambled control microRNA. *P<0.05 versus SC-AdLacZ, ^#P<0.05 versus miR-204-AdLacZ. A, acetylcholine; AB, antibiotics; eSirt1^−/−, mice conditionally lacking endothelial Sirt1; IF, immunofluorescence; NFR, nuclear fast red; NPD, normal pellet diet; PE, phenylephrine; R, stoppage of antibiotics and recolonization; Sirt1^fl/fl, control Sirt1 floxed mice. n represents the number of aortic rings in h–l. Independent sample t-test was used. Data shown as mean and error bar represents s.e.m.

Figure 3. Antibiotics rescue endothelial Stat3 and Sirt1 in high-fat diet-induced vascular dysfunction.

(a–c) Suppression of gut microbiota with broad-spectrum antibiotics inhibits HFD-induced aortic miR-204. (a) qPCR for mature and precursor miR-204 in mouse aortas. n=4 for NPD and HFD-AB-R, n=6 for HFD and HFD-AB. (b) In situ hybridization for miR-204 in mouse aortas (blue, magnification × 100, Scale bar; 20 μm). (c) Quantification of miR-204 in b. n=5 images from different regions of aortic sections. (d–f) Suppression of gut microbiome with antibiotics prevents HFD-induced downregulation of aortic Sirt1 and Stat3 signalling. (d) Immunofluorescence for Sirt1 in mouse aortic sections (magnification × 63, Scale bar; 20 μm). (e) Quantification of total aortic (n=4 sections) and endothelial (n=15 endothelial cells) Sirt1 immunostaining in d. (f) Western blot for Sirt1 and phospho-Stat3 in mouse aortas. Immunoblots were done with pooled aortic samples; n=4 for NPD and HFD-AB-R and n=6 for HFD and HFD-AB. (g) Suppression of gut microbiome rescues HFD-induced decrease in total aortic NO_x [nitrate+nitrite]. NPD: n=9, HFD: n=6, HFD-AB: n=6. (h,i) Suppression of gut microbiota rescues HFD-induced reduction of endothelium-dependent vasorelaxation and nitric oxide bioavailability. NPD: n=10, HFD: n=12, HFD-AB: n=10, HFD-AB-R: n=10. (j) Suppression of gut microbiota does not rescue HFD-induced impairment of endothelium-dependent vasorelaxation in eSirt1^−/− mice. eSirt1^−/− HFD: n=6, eSirt1^−/−HFD-AB: n=12. *P<0.05 versus NPD. ^#P<0.05 versus HFD. A, acetylcholine; AB, antibiotics; eSirt1^−/−, mice conditionally lacking endothelial Sirt1; HFD, high-fat diet; NPD, normal pellet diet; PE, phenylephrine; R, stoppage of antibiotics and recolonization. n represents the number of aortic rings in h–j. IF, immunofluorescence. Independent sample t-test was used. Data shown as mean and error bar represents s.e.m.

Figure 4. MiR-204 mediates high-fat diet-induced vascular dysfunction and inflammation.

(a) Systemic delivery of miR-204 inhibitor in HFD-fed mice (a) prevents upregulation of aortic miR-204, (b) downregulation of aortic Sirt1 and (c,d) rescues endothelium-dependent vasorelaxation and bioavailable vascular NO. n=4 for NPD-SC, n=6 for HFD-SC and HFD-204-I in a; n=4 for NPD-SC, n=6 for HFD-SC and HFD-204-I in b; NPD-SC: n=15, HFD-SC: n=16, HFD-204-I: n=18 in c,d. n represents number of aortic rings. (e,f) Systemic in vivo delivery of miR-204 inhibitor suppresses HFD-induced vascular inflammation. Expression of macrophage marker f4/80 was assessed by immunoblot in aorta (e) and by immunohistochemistry (f) in aortic roots of mice (magnification × 4 or × 100; scale bars, 500 μm (× 4), 20 μm (× 100)). Adjacent sections of aortic root sections were stained with haematoxylin and eosin (magnification × 4 and 100). n=4 for NPD-SC, n=6 for HFD-SC and HFD-204-I, *P<0.05 versus NPD-SC. ^#P<0.05 versus HFD-SC. HFD, high-fat diet; NPD, normal pellet diet; SC, scrambled control microRNA; 204-I, miR-204 inhibitor. (g–i) Inhibition of miR-204 with HFD feeding improves endothelium-dependent vasorelaxation to a lesser degree in eSirt1^−/− mice compared with wild-type mice. (g) Aortic expression of miR-204 in eSirt1^−/− mice with HFD and with miR-204 inhibitor infusion. RT–PCR was done with pooled aortic samples. (h) Systemic delivery of miR-204 inhibitor partially improves endothelium-dependent vasorelaxation induced in eSirt1^−/− mice on HFD. n represents the number of aortic rings. NPD-SC: n=13, HFD-SC: n=12, HFD-204-I: n=9. (i) Per cent impairment in endothelium-dependent vascular relaxation in c,h. (j) Ex vivo inhibition of miR-204 rescues HFD-induced impairment of endothelium-dependent vasorelaxation. n represents the number of aortic rings. NPD-SC: n=5, HFD-SC: n=10, HFD-204-I: n=10, *P<0.05 versus NPD-SC. ^#P<0.05 versus HFD-SC. A, acetylcholine; eSirt1^−/−, mice conditionally lacking endothelial Sirt1; HFD, high-fat diet; L, lumen; NPD, normal pellet diet; PE, phenylephrine. Independent sample t-test was used. Data shown as mean and error bar represents s.e.m.

Figure 5. Remote regulation of vascular endothelial function by the microbiome.

Gut microbiota-regulated second messengers in the systemic circulation induce vascular miR-204 expression which, in turn, promotes endothelial dysfunction by targeting Sirt1.