Rap1b in smooth muscle and endothelium is required for maintenance of vascular tone and normal blood pressure

Abstract

Objective: Small GTPase Ras-related protein 1 (Rap1b) controls several basic cellular phenomena, and its deletion in mice leads to several cardiovascular defects, including impaired adhesion of blood cells and defective angiogenesis. We found that Rap1b(-/-) mice develop cardiac hypertrophy and hypertension. Therefore, we examined the function of Rap1b in regulation of blood pressure.

Approach and results: Rap1b(-/-) mice developed cardiac hypertrophy and elevated blood pressure, but maintained a normal heart rate. Correcting elevated blood pressure with losartan, an angiotensin II type 1 receptor antagonist, alleviated cardiac hypertrophy in Rap1b(-/-) mice, suggesting a possibility that cardiac hypertrophy develops secondary to hypertension. The indices of renal function and plasma renin activity were normal in Rap1b(-/-) mice. Ex vivo, we examined whether the effect of Rap1b deletion on smooth muscle-mediated vessel contraction and endothelium-dependent vessel dilation, 2 major mechanisms controlling basal vascular tone, was the basis for the hypertension. We found increased contractility on stimulation with a thromboxane analog or angiotensin II or phenylephrine along with increased inhibitory phosphorylation of myosin phosphatase under basal conditions consistent with elevated basal tone and the observed hypertension. Cyclic adenosine monophosphate-dependent relaxation in response to Rap1 activator, Epac, was decreased in vessels from Rap1b(-/-) mice. Defective endothelial release of dilatory nitric oxide in response to elevated blood flow leads to hypertension. We found that nitric oxide-dependent vasodilation was significantly inhibited in Rap1b-deficient vessels.

Conclusions: This is the first report to indicate that Rap1b in both smooth muscle and endothelium plays a key role in maintaining blood pressure by controlling normal vascular tone.

Keywords: relaxation; signal transduction; vasodilation.

Figure 1

Pathological cardiac hypertrophy in Rap1b^−/− mice. (A) Representative images of 17-month-old Rap1b^−/− hearts compared to age-matched wild-type (WT, Rap1b^+/+) control. Scale bar, 2 mm. (B) Quantification of heart to body weight ratios of WT and Rap1b^−/− mice. (n≥10). P<0.0001. (C) Representative M-mode echocardiogram from 6–10 month old Rap1b^−/− and WT mice; measurements acquired from the echocardiography are presented in Table 1. (D) Representative images (top) and quantification (bottom) of cardiomyocyte size in WT and Rap1b^−/− ventricular heart cross section stained with Periodic acid-Schiff reagent. (n=5). Scale bar, 150 μm. (E, F) Increased fibrosis and inflammation in Rap1b^−/− hearts. (E) Representative images of Masson trichrome-stained WT and Rap1b^−/− heart sections (top) and quantification of cardiac fibrosis in WT (n = 5) and Rap1b^−/− (n=11) hearts. Scale bar, 2 mm. (F) Representative images of F4/80 IHC stained WT and Rap1b^−/− heart sections (top). Quantification of invading monocytes/macrophages assessed by F4/80 immunolocalization in heart sections (bottom, n≥3). Values are means ± s.e.m.

Figure 2

(A–C) Adult Rap1b^−/− mice are hypertensive. Absolute values of telemetrically-measured arterial blood pressure in 6–10 month old (A) and older (>12 month old, B) WT (n=5) and Rap1b^−/− mice. (n≥4). (C) Representative blood pressure traces of age-matched WT and Rap1b^−/− mice. MAP: mean arterial pressure. (D–E) Competitive inhibitor of angiotensin II type I receptor (AT₁R), losartan, corrects hypertension and cardiac hypertrophy phenotype in Rap1b^−/− mice. (D) Systolic blood pressure in 8–12 week old Rap1b^−/− mice following chronic losartan treatment (+L; n=5) is lowered to that of WT controls (n=3). (E) Quantification of heart to body weight ratios of WT and Rap1b^−/− mice without or with chronic losartan treatment (+L) (n≥6 of each genotype per group). Losartan treatment ameliorates increased heart to body ratio in Rap1b^−/− mice to the level observed in WT mice. (F–G) Anti-hypertensive drug hydralazine corrects hypertension but not cardiac hypertrophy in Rap1b^−/− mice. (F) Systolic blood pressure in 8–12 week old Rap1b^−/− mice following chronic hydralazine treatment (+L; n=7) is lowered to that of WT controls (n=5). (G) Quantification of heart to body weight ratios of WT and Rap1b^−/− mice without or with chronic hydralazine treatment (n≥4). Values are means ± s.e.m.

Figure 3

Renal function is preserved in Rap1b^−/− mice. (A) Mean urine output per day in Rap1b^−/− and WT control mice (n=4) is within normal range of 0.5–1.0 ml/day for mice . (B) Kidney to body weight ratio is comparable between Rap1b^−/− and WT mice (n≥9). Urine microalbumin (C) and urine creatinine (D) are not significantly altered in Raplb^−/− mice (n=4), and within normal range of up to 400 μg/day . (E) Unchanged urine microalbumin to creatinine ratio suggests that renal function is not altered in Rap1b^−/− mice. Plasma renin activity (F, n=4) and plasma creatinine (G, n≥5) were unchanged in Rap1b^−/− mice, compared to WT mice, and within normal range of 0.1–0.22 mg/dL. Values are means ± s.e.m.

Figure 4

Normal Rap1b^−/− kidney histology. (A) Representative low power images of 6-month-old Rap1b^−/− kidney sections compared to age-matched, wild-type controls (WT), show no gross morphology changes in Rap1b^−/− kidneys. Scale bar, 2 mm. (B) Representative high power images of cortex, outer medulla and medulla in WT and Rap1b^−/− kidney section stained with Masson-trichrome reagent and counter-stained with H & E, suggest that Rap1b^−/− kidney morphology is not significantly different as compared to WT. Arrow indicates RBCs and asterisk - glomerulus. Scale bar, 150 μm. (n≥8).

Figure 5

Contractile responses to the thromboxane analog, U46619, angiotensin II (AngII) or phenylephrine and the relaxant effect of 8-pCPT-2′-O-Me-cAMP (007), a cAMP analog specific for Rap1GEF, Epac, in endothelium-denuded aortae (A,B). (A) Dose-dependent contraction in WT and Rap1b^−/− aortae in the absence or presence of 007. Amplitude of the vessel contraction measured in muscle strips of the same cross sectional area. Plots are calculated means ± s.e.m.; lines are fit plots. (n=5). Statistical analysis of data shown in panel A is included in Supplemental Table I. (B) Bar graph shows maximal contractile force (mN) induced by 500nM U46619 ± 100 μM 007. (n=5). (C,D,F) Contraction of WT and Rap1b^−/− aortae in response to high K⁺(C), 25nM angiotensin II (D) or phenylephrine (F). (E) Immunoblot of AT₁R expression in WT and Rap1b^−/ aortae, typical results are shown.

Figure 6

(A) Actin and myosin RLC₂₀ expression in WT and Rap1b^−/− aortas. (n=7). (B) Immunoblot analysis and quantification of basal phosphorylation of MYPT1/Thr853, normalized to total MYPT1, and RLC₂₀, normalized to actin, in WT and Rap1b^−/− aortae (n=7). (C) Immunoblot analysis and densitometric quantification of RLC₂₀ phosphorylation over a 5 min time course following treatment with 500nM U46619 in WT or Rap1b^−/− aortic strips (n=3). WT values at 0 sec are taken as 1. P<0.05. Values are means ± s.e.m. (D) Proposed mechanism of Rap1-induced SM relaxation. Basal or elevated cAMP, via Epac, activates Rap1 (top), which, through an unknown RhoGAP, inhibits RhoA/ROCK activity. This results in an increase in MLCP activity due to a decrease in ROCK-mediated myosin light chain phosphatase (MLCP) inhibitory phosphorylation (“dis-inhibition”), dephosphorylation of RLC₂₀ and relaxation or decreased basal tone. When Rap1 is inactive (bottom) or absent (Rap1b^−/− vessels) Rho activity is not suppressed resulting in increased MLCP inhibitory phosphorylation (p-MLCP), increased RLC₂₀ phosphorylation and increased basal SM contraction. Activated signaling molecules in each scenario are highlighted in green.

Figure 7

Endothelium-dependent relaxation of U46619-preconstricted aortae and activation of eNOS. Dose-dependent relaxation of WT (grey line, n=11) Rap1b^−/− (black line, n=10) and Tie2-Cre^+/0; Rap1b^f/f (Rap1b-ECKO, black dashed line, n=6) aortic rings in response to acetylcholine (Ach, A) or sodium nitroprusside (SNP), an NO donor, in the presence of L-NAME (n=8, 6 and 4, respectively; B). Relaxation is expressed as a fold maximal dilation relative to U46619 constriction. Plotted are calculated means ± s.e.m.; lines are fit plots. (C) Immunoblot and quantification of VEGF-induced phosphorylation of eNOS/Ser1177, normalized to actin in HUVECs transfected with control, scrambled siRNA or Rap1b siRNA. Fold change vs. unstimulated control is shown (n=3).