Angiotensinogen Exerts Effects Independent of Angiotensin II

Abstract

Objective: This study determined whether angiotensinogen (AGT) has angiotensin II-independent effects using multiple genetic and pharmacological manipulations.

Approach and results: All study mice were in low-density lipoprotein receptor -/- background and fed a saturated fat-enriched diet. In mice with floxed alleles and a neomycin cassette in intron 2 of the AGT gene (hypoAGT mice), plasma AGT concentrations were >90% lower compared with their wild-type littermates. HypoAGT mice had lower systolic blood pressure, less atherosclerosis, and diminished body weight gain and liver steatosis. Low plasma AGT concentrations and all phenotypes were recapitulated in mice with hepatocyte-specific deficiency of AGT or pharmacological inhibition of AGT by antisense oligonucleotide administration. In contrast, inhibition of AGT cleavage by a renin inhibitor, aliskiren, failed to alter body weight gain and liver steatosis in low-density lipoprotein receptor -/- mice. In mice with established adiposity, administration of AGT antisense oligonucleotide versus aliskiren led to equivalent reductions of systolic blood pressure and atherosclerosis. AGT antisense oligonucleotide administration ceased body weight gain and further reduced body weight, whereas aliskiren did not affect body weight gain during continuous saturated fat-enriched diet feeding. Structural comparisons of AGT proteins in zebrafish, mouse, rat, and human revealed 4 highly conserved sequences within the des(angiotensin I)AGT domain. des(angiotensin I)AGT, through adeno-associated viral infection in hepatocyte-specific AGT-deficient mice, increased body weight gain and liver steatosis, but did not affect atherosclerosis.

Conclusions: AGT contributes to body weight gain and liver steatosis through functions of the des(angiotensin I)AGT domain, which are independent of angiotensin II production.

Keywords: angiotensinogen; atherosclerosis; blood pressure; liver steatosis; obesity.

Figure 1. Global reductions of AGT decreased blood pressure, atherosclerosis, body weight gain and liver steatosis in LDL receptor -/- mice

All mice were male and fed a Western diet for 12 weeks. (A) Plasma AGT concentrations were measured by ELISA. N = 6-7/group. * P = 0.001 by Mann-Whitney Rank Sum Test. (B) Plasma renin concentrations were measured by radioimmunoassay. N = 6-7/group. * P < 0.001 by Student's t-test. (C) Systolic blood pressure was measured using a tail-cuff system. N = 15-19/group. * P < 0.001 by Mann-Whitney Rank Sum Test. (D) Atherosclerotic lesions in the aortic arch region were quantified as percent intimal area covered by lesions. N = 15-19/group. * P < 0.001 by Mann-Whitney Rank Sum Test. (E) Body weights were measured weekly. N = 6-8/group. (F) Body composition was determined by EchoMRI prior to (week 0) and during week 12 of Western diet feeding. N = 6-8/group. * P < 0.001 for fat mass during Western diet feeding by two way repeated measures ANOVA with Holm-Sidak method. (G) Liver weights were measured at termination. N = 6-8/group. * P < 0.01 by Student's t-test. (H) Liver triglyceride content was measured using mass spectrometry after termination. N = 6-8/group. * P = 0.001 by Mann-Whitney Rank Sum Test. (I) H&E staining of liver.

Figure 2. Hepatocyte-specific deficiency of AGT decreased blood pressure, atherosclerosis, body weight gain and liver steatosis

All mice were male fed a Western diet for 12 weeks. (A) Plasma AGT concentrations were measured by ELISA. N = 8-9/group. * P < 0.001 by Mann-Whitney Rank Sum Test. (B) Plasma renin concentrations were measured by radioimmunoassay. N = 6-9/group. * P = 0.002 by Mann-Whitney Rank Sum Test. (C) Systolic blood pressure was measured using a tail-cuff system. N = 16-19/group. * P < 0.001 by Student's t-test. (D) Atherosclerotic lesions in the aortic arch region were quantified as percent intimal area covered by lesions. N = 16-17/group. * P < 0.001 by Mann-Whitney Rank Sum Test. (E) Body weights were measured weekly. N = 8-9/group. (F) Body composition was determined by EchoMRI prior to (week 0) and during week 12 of Western diet feeding. N = 8-9/group. * P = 0.04 and P < 0.001 for lean and fat masses, respectively, during Western diet feeding by two way repeated measures ANOVA with Holm-Sidak method. (G) Liver weights were measured at termination. N = 8-9/group. * P = 0.001 by Student's t-test. (H) Liver triglyceride contents were measured using mass spectrometry. N = 7-8/group. * P = 0.007 by Student's ***t-test.

Figure 3. Reduction of AGT abundance by second generation ASOs decreased blood pressure, atherosclerosis, and body weight gain in LDL receptor -/- mice

Male mice were fed a Western diet for 12 weeks and administered 50 mg/kg of ASOs once weekly via intraperitoneal injection. (A) Plasma AGT concentrations were measured by ELISA. N = 6-9/group. * P < 0.001 comparing AGT-1 and -2 to control ASO (CON) by one way ANOVA with Holm-Sidak method. “AGT-1” is AGT ASO # 261333, and “AGT-2” is AGT ASO # 487022. (B) Plasma renin concentrations were measured by radioimmunoassay. N = 8-10/group. * P < 0.001 comparing AGT-1 and -2 to CON by Kruskal-Wallis one way ANOVA on Ranks. (C) Systolic blood pressure was measured using a tail-cuff system. N = 9-10/group. * P < 0.001 comparing AGT-1 and 2 to CON by one way ANOVA with Holm-Sidak method. (D) Atherosclerotic lesions in the aortic arch region were quantified as percent intimal area covered by lesions. N = 9-10/group. * P = 0.009 comparing AGT-1 and 2 to CON by one way ANOVA with Holm-Sidak method. (E) Body weights were measured weekly. (F) Body composition was determined by EchoMRI during 12 weeks of Western diet feeding. N = 10/group. * P < 0.001 comparing AGT-1 and 2 to CON for fat mass, respectively, by one way ANOVA with Holm-Sidak method. (G) Representative images of mice injected with control versus AGT ASO (AGT-1). (H) H&E staining of epididymal fat in mice injected with control versus AGT ASO (AGT-1).

Figure 4. Renin inhibition did not affect plasma AGT concentrations, body weight gain, or liver steatosis in LDL receptor -/- mice

Male mice were fed a Western diet and infused with aliskiren subcutaneously for 12 weeks. (A) Plasma AGT concentrations were measured by ELISA. N = 7-8/group. P = 0.6 by Student's t-test. (B) Body weights were measured weekly. N = 15/group. (C) Liver triglyceride contents were measured using mass spectrometry. N = 4-5/group. P = 0.5 by Student's t-test.

Figure 5. Inhibition of AGT, not renin, regressed pre-existing obesity

Systolic blood pressure was measured using a tail-cuff system on mice injected with control or AGT ASO (AGT-1: # 261333) (A and B) or infused with vehicle or aliskiren (C and D). N = 12-24/group. Systolic blood pressure changes were calculated by comparing measurements during drug administration with measurements prior to drug administration. * P = 0.04 by Mann-Whitney Rank Sum Test for comparisons in (A), and * P < 0.001 by Student's t-test for comparisons in (C). Gain of body weight was determined by comparing weekly body weight measured weekly with baseline body weight prior to Western diet feeding (B and D).

Figure 6. Structural analysis of highly conserved regions in des(AngI)AGT and its biological effect

(A) Conserved structural features were analyzed based on bioinformatic analyses of AGT gene sequences from human, rat, mouse, and zebrafish. Four highly conserved surface regions (Core, renin interacting, beta-sheet face, and loop) were identified in des(AngI)AGT domain. (B) Plasma renin concentrations were measured by radioimmunoassay. N = 10-14/group. * P < 0.001 versus hepAGT+/+ mice injected with null AAV by Kruskal-Wallis one way ANOVA on rank method. (C) Atherosclerotic lesions in the aortic arch region were quantified as percent intimal area covered by lesions. N = 9-27/group. * P < 0.001 versus hepAGT+/+ mice injected with null AAV by one way ANOVA with Holm-Sidak method. Body weights (D) and liver weights (E) were measured at termination. N = 10-27/group. * P < 0.001 versus hepAGT-/- mice injected with null AAV by Kruskal-Wallis one way ANOVA on rank method. (F) Liver triglyceride contents were measured using mass spectrometry after termination. N = 4-5/group. * P = 0.005 versus hepAGT-/- mice infected with a null AAV by one way ANOVA with Holm-Sidak method after square root transformation.