Gut microbiota dependent trimethylamine N-oxide aggravates angiotensin II-induced hypertension

Abstract

Gut microbiota produce Trimethylamine N-oxide (TMAO) by metabolizing dietary phosphatidylcholine, choline, l-carnitine and betaine. TMAO is implicated in the pathogenesis of chronic kidney disease (CKD), diabetes, obesity and atherosclerosis. We test, whether TMAO augments angiotensin II (Ang II)-induced vasoconstriction and hence promotes Ang II-induced hypertension. Plasma TMAO levels were indeed elevated in hypertensive patients, thus the potential pathways by which TMAO mediates these effects were explored. Ang II (400 ng/kg^-1min^-1) was chronically infused for 14 days via osmotic minipumps in C57Bl/6 mice. TMAO (1%) or antibiotics were given via drinking water. Vasoconstriction of renal afferent arterioles and mesenteric arteries were assessed by microperfusion and wire myograph, respectively. In Ang II-induced hypertensive mice, TMAO elevated systolic blood pressure and caused vasoconstriction, which was alleviated by antibiotics. TMAO enhanced the Ang II-induced acute pressor responses (12.2 ± 1.9 versus 20.6 ± 1.4 mmHg; P < 0.05) and vasoconstriction (32.3 ± 2.6 versus 55.9 ± 7.0%, P < 0.001). Ang II-induced intracellular Ca²⁺ release in afferent arterioles (147 ± 7 versus 234 ± 26%; P < 0.001) and mouse vascular smooth muscle cells (VSMC, 123 ± 3 versus 157 ± 9%; P < 0.001) increased by TMAO treatment. Preincubation of VSMC with TMAO activated the PERK/ROS/CaMKII/PLCβ3 pathway. Pharmacological inhibition of PERK, ROS, CaMKII and PLCβ3 impaired the effect of TMAO on Ca²⁺ release. Thus, TMAO facilitates Ang II-induced vasoconstriction, thereby promoting Ang II-induced hypertension, which involves the PERK/ROS/CaMKII/PLCβ3 axis.

Keywords: Afferent arteriole; Angiotensin II; Blood pressure; Calcium; Trimethylamine N-oxide.

Graphical abstract

Fig. 1

Chronic administration of TMAO aggravates angiotensin II (Ang II)-induced hypertension and impairs renal glomerular filtration rate (GFR). (A) Systolic blood pressure (SBP) in mice delivered saline, Ang II (400 ng/kg⁻¹min⁻¹), TMAO (1%) and/or an antibiotic cocktail (n = 4–7; *P < 0.05 vs Control, ^#P < 0.05 vs Ang II). (B) Time course of glomerular filtration rate (GFR) in mice treated with TMAO (n = 5–6; *P < 0.05 vs 0). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (A), one-way ANOVA followed by Turkey's post hoc test (B).

Fig. 2

Chronic administration of TMAO augments Ang II-induced vasoconstriction of mesenteric artery and afferent arterioles via activation of phospholipase C (PLC). Vasoconstriction in afferent arteriole (Af) or mesenteric artery were assessed by microperfusion or wire myograph. (A) Representative images of the isolated perfused Af before (Basal) and during administration of 10⁻⁹ and 10⁻⁶ mol/L Ang II. Concentration-response curves to Ang II of Af (B) or mesenteric artery (C) from TMAO, Antibiotics or vehicle-treated mice infused with saline or Ang II (n = 4–14; *P < 0.05 vs Control, ^#P < 0.05 vs Ang II, ^$P < 0.05 vs Ang II+ Antibiotics). (D) Mesenteric artery response to Ang II in the presence of PLC inhibitor U73122 (n = 8). (E) Mesenteric artery contraction response to phenylephrine (PE). (F) Effect of U73122 on PE induced vasoconstriction of mesenteric artery (n = 6). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (B–F).

Fig. 3

Acute administration of TMAO increases pressor response and vasoconstriction to Ang II. (A) Direct effect of TMAO on SBP (n = 4–5; *P < 0.05 vs NaCl). (B–C) Direct effect of TMAO on vascular tone in Af and mesenteric artery (n = 4–5; *P < 0.05 vs 0). (D) The changes of SBP during two intravenous bolus injections of 0.25 μg/kg Ang II. Five minutes after the first injection of Ang II, TMAO or saline was administered (n = 6–7; *P < 0.05 vs NaCl). (E) The changes of Ang II caused vasoconstriction in the absence or presence of TMAO (n = 6–7, *P < 0.05 vs Control). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (A, D and E), one-way ANOVA followed by Turkey's post hoc test (B–C). 1st, First Application; 2nd, Second Application.

Fig. 4

TMAO activates the PERK/elF2α/Chop/ERO1-α pathway. (A) Effect of 1 h preincubation with TMAO (250 μM) on the phosphorylation of PERK in Af. (B–C) The levels of p-PERK, PERK, p-elF2α, elF2α, CHOP and ERO1-α were detected by western blotting after incubation of mouse vascular smooth muscle cell (VSMC) with various concentrations of TMAO for 24 h or a fixed concentration of TMAO (100 μM) for different time periods (n = 3; *P < 0.05 vs 0). Statistical differences were calculated by one-way ANOVA followed by Turkey's post hoc test (B–C).

Fig. 5

Reactive oxygen species (ROS) are involved in regulating the effect of TMAO on Ang II-induced pressor responses and vasoconstriction. The changes of the Ang II-induced pressor response (A) and vasoconstriction (B) in the presence of cytoplasmic ROS scavenger Tempol or Tempol+TMAO (n = 4; *P < 0.05 vs NaCl+Tempol or Control+Tempol). (C–D) Afs were incubated with TMAO (250 μM) or CCT020312 (10 μM) for 1 h and then cytoplasmic superoxide or mitochondrial superoxide were measured by staining with DHE or MitoSOX, respectively (n = 3–4; *P < 0.05 vs Control). (E) Changes in cytoplasmic superoxide and mitochondrial superoxide levels in VSMC incubated with TMAO, PERK inhibitor GSK, mitochondrial ROS scavenger MitoTempo, cytoplasmic ROS scavenger Tempol or Ang II (n = 6–18; *P < 0.05 vs Control, $P < 0.05 vs Control+Tempol, #P < 0.05 vs Control+Ang II). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (A–B), one-way ANOVA followed by Turkey's post hoc test (C–E). 1st, First Application; 2nd, Second Application; GSK, GSK2606414; CCT, CCT020312.

Fig. 6

Calmodulin-dependent protein kinase II (CaMKII) is involved in regulating the effect of TMAO on Ang II-induced pressor response and vasoconstriction. (A–B) The changes of Ang II-induced pressor response and vasoconstriction in the continuous presence of CaMKII inhibitor KN-93 or KN-93 + TMAO (n = 4). (C–D) The levels of Ox-CaMKII, p-CaMKII and CaMKII were detected by western blotting after incubation of VSMC with various concentrations of TMAO for 24 h or a fixed concentration of TMAO (100 μM) for different time periods (n = 3; *P < 0.05 vs 0). Statistical differences were determined by two-way ANOVA followed by Dunnett's post hoc test (A–B), one-way ANOVA followed by Turkey's post hoc test (C–D). 1st, First Application; 2nd, Second Application; GSK, GSK2606414.

Fig. 7

Activation of the PLCβ3/IP₃/Ca²⁺pathway by TMAO promotes Ang II-induced vasoconstriction. (A) Effect of U73122 treatment on Ang II-induced vasoconstriction in the absence or presence of TMAO (n = 4; *P < 0.05 vs Control+U73122). (B–C) Effect of TMAO on Ang II-induced PLC activation and IP₃ formation in VSMC (n = 6–12; *P < 0.05 vs Control, ^#P < 0.05 vs Control+Ang II). (D–E) The levels of p-PLCβ3 and p-PKC were detected by western blotting after incubation of VSMC with various concentrations of TMAO for 24 h or a fixed concentration of TMAO for different time periods (n = 3; *P < 0.05 vs 0). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (A), one-way ANOVA followed by Turkey's post hoc test (B–E).

Fig. 8

TMAO promotes Ang II-evoked intracellular Ca2+ release. (A) Representative images of the isolated perfused Afs loaded with fluo-4 AM for measurement of intracellular Ca²⁺. In the presence of TMAO, Ang II triggered intracellular Ca²⁺ release was significantly increased, which was blocked by U73122 and Tempol (n = 3–5; *P < 0.05 vs Control). (B) Representative images and summarized data showing that TMAO enhances intracellular Ca²⁺ release of VSMC in response to Ang II (0.1 μM), which was impaired by GSK, Tempol, KN-93, U73122 and BAPTA-AM (n = 4–8; *P < 0.05 vs Control, ^#P < 0.05 vs TMAO). Statistical differences were calculated by two-way ANOVA followed by Dunnett's post hoc test (A), one-way ANOVA followed by Turkey's post hoc test (B). GSK, GSK2606414.

Fig. 9

Plasma TMAO is correlated with SBP, SOD activity, creatinine and BUN in normotensive and hypertensive adults. Plasma levels of TMAO (A), SOD activity (C), creatinine (E) and BUN (G) in normotensive (n = 32) or hypertensive patients (n = 37) (*P < 0.05 vs Normotensive). Correlation analysis of plasma TMAO with SBP (B), SOD activity (D), creatinine (F) and BUN (H). Statistical differences were calculated by the paired T-test (A, C, E and G), The correlation between 2 parameters was evaluated by Pearson correlation test (B, D, F and H).