Angiotensin II triggers release of neutrophil extracellular traps, linking thromboinflammation with essential hypertension

Abstract

Innate immunity and chronic inflammation are involved in atherosclerosis and atherothrombosis, leading to target organ damage in essential hypertension (EH). However, the role of neutrophils in EH is still elusive. We investigated the association between angiotensin II (Ang II) and neutrophil extracellular traps (NETs) in pathogenesis of EH. Plasma samples, kidney biopsies, and surgical specimens of abdominal aortic aneurysms (AAAs) from patients with EH were used. Cell-based assays, NETs/human aortic endothelial cell cocultures, and in situ studies were performed. Increased plasma levels of NETs and tissue factor (TF) activity were detected in untreated, newly diagnosed patients with EH. Stimulation of control neutrophils with plasma from patients with untreated EH generated TF-enriched NETs promoting endothelial collagen production. Ang II induced NETosis in vitro via an ROS/peptidylarginine deiminase type 4 and autophagy-dependent pathway. Circulating NETs and thrombin generation levels were reduced substantially in patients with EH starting treatment with Ang II receptor blockers, whereas their plasma was unable to trigger procoagulant NETs. Moreover, TF-bearing NETotic neutrophils/remnants accumulated in sites of interstitial renal fibrosis and in the subendothelial layer of AAAs. These data reveal the important pathogenic role of an Ang II/ROS/NET/TF axis in EH, linking thromboinflammation with endothelial dysfunction and fibrosis.

Keywords: Hypertension; Immunology; Inflammation; Neutrophils; Thrombin.

Figure 1. Markers of NETs are detected both in plasma of patients with EH and in control neutrophils treated with EH plasma.

(A) MPO-DNA complex levels and (B) CitH3 levels representing NET release in plasma from healthy individuals (controls, n = 26) and patients with EH (n = 55). For A and B, lines represent means accompanied by their ±95% CI, Student’s t test (2 tailed). (C) Correlation between MPO-DNA levels and CitH3 levels in 2 patients, Pearson’s correlation test. (D) Fluorescence microscopy images showing NE/CitH3 staining (blue: DAPI, green: NE, red: CitH3, original magnification, 1000×) and (E) percentage of NET release as assessed by immunofluorescence, in control neutrophils incubated with EH plasma and inhibited using diphenyleneiodonium (DPI; NADPH oxidase 2 inhibitor) or Cl-amidine (pan–protein arginine deiminase [pan-PAD] inhibitor). For D, a representative example of 6 independent experiments is shown. (F) MPO-DNA complex levels in NETs isolated from control neutrophils treated with EH plasma and inhibited with DPI or Cl-amidine. Red and green dots indicate values yielded from control neutrophils that had been incubated with EH plasma samples with higher or lower levels of NET markers, respectively. For D–F, ionomycin-stimulated neutrophils were used as positive controls. For E and F, data are from 6 independent experiments (mean ± SD, Friedman’s test). All conditions were compared to controls/untreated (statistically significant: P < 0.05; NS: not significant).

Figure 2. NETs in EH express TF.

(A) Thrombin-antithrombin (TAT) complex levels in plasma from healthy individuals (controls, n = 26) and EH patients (n = 55). Lines represent means accompanied by their ±95% CI, Student’s t test (2 tailed). (B) Correlation between CitH3 representing NET release and TAT levels in EH patients, Pearson’s correlation test. TF expression in control neutrophils treated with EH plasma as assessed by (C) qPCR or (D) in-cell ELISA. For C, GAPDH was used to normalize gene expression. For C and D, Wilcoxon’s test for paired samples was used. (E) Fluorescence microscopy images showing TF/NE staining (blue, DAPI; green, TF; red, NE; original magnification, 1000×) in control neutrophils incubated with EH plasma. A representative example of 6 independent experiments is shown. (F and G) TAT levels and TF activity in in vitro–isolated NET structures, respectively. NETs were obtained by control neutrophils incubated with EH plasma, and subsequently inhibited by DNase I or anti-TF neutralizing antibody, Friedman’s test. In F, red and green dots indicate values yielded by incubation of control neutrophils with EH plasma samples that had higher or lower TAT levels, respectively. For C, D, F, and G, data are from 6 independent experiments (mean ± SD). All conditions were compared with controls/untreated (statistically significant: P < 0.05).

Figure 3. Ang II induces NET formation.

(A–F) Fluorescence microscopy images showing NE/CitH3 staining (blue, DAPI; green, TF; red, NE; original magnification, 1000×) in control neutrophils incubated with 0.1 nM Ang II and inhibited with irbesartan (ARB), DPI (NADPH oxidase 2 inhibitor), wortmannin (early-stage autophagy inhibitor), or Cl-amidine (pan-PAD inhibitor). A representative example of 6 independent experiments is shown. (G) Percentage of NET release assessed by immunofluorescence and (H) MPO-DNA complex levels in in vitro–isolated NET structures. For G and H, data are from 6 independent experiments (mean ± SD), Friedman’s test. Inhibitions were performed as described in A–F. All conditions were compared with untreated (statistically significant: P < 0.05).

Figure 4. ARBs diminish NETs and TAT activity in EH.

Paired analysis of (A) CitH3 levels and (B) TAT complex levels in plasma, which was obtained from the same EH patients before and under treatment with ARBs (n = 12). For A and B, lines represent means accompanied by their ±95% CI, Wilcoxon’s test for paired samples. (C) MPO-DNA complex levels in NETs isolated from control neutrophils stimulated with plasma obtained from EH patients before (EH-plasma) and under treatment with ARB (ARB EH plasma). (D) TAT levels in control plasma incubated with NET structures that were isolated from control neutrophils treated with EH plasma or ARB EH plasma, as described in C. For C and D, data are from 6 independent experiments (mean ± SD), Friedman’s test. All conditions were compared with untreated (statistically significant: P < 0.05).

Figure 5. HAoECs acquire a profibrotic phenotype upon incubation with the thromboinflammatory NETs of EH.

(A) Relative fold expression of mRNA assessed by qPCR and (B) surface protein expression assessed by in-cell ELISA for intercellular adhesion molecule 1 (ICAM1) and vascular cell adhesion molecule 1 (VCAM1). HAoECs were incubated with NETs released from neutrophils upon stimulation with plasma from EH patients before (EH-NETs) and after treatment with ARBs (ARB EH-NETs). DNase I was used to dismantle NETs. (C) Relative fold expression of mRNA for connective tissue growth factor (CCN2), (D) CCN2 protein expression assessed by in-cell ELISA, and (E) release of collagen in HAoECs incubated with NETs, as described in A and B. To hinder TF/thrombin axis, HAoECs were pretreated with FLLRN (PAR1 receptor inhibitor) or EH-NETs preincubated with a neutralizing antibody against TF. For A and C, RPL13A was used to normalize gene expression. For A–E, data are from 4 independent experiments (mean ± SD), Friedman’s test. All conditions were compared with untreated (statistically significant: P < 0.05).

Figure 6. NETotic neutrophils expressing TF are identified in kidney biopsies and AAA specimens from patients with EH.

(A) NETotic neutrophils/remnants, visualized in renal specimens from a patient with hypertensive nephropathy by costaining with NE and CitH3 (confocal microscopy: blue, DAPI; green, NE; red, CitH3; original magnification, 630×), express TF (confocal microscopy: blue, DAPI; green, TF; red, NE; original magnification, 630×). White arrowheads indicate the renal tubules (either proximal or distant). Renal biopsy is characterized by interstitial fibrosis, as assessed by (B) Masson’s trichrome staining and (C) hematoxylin & eosin staining (light microscopy, original magnification, 100×). For A and C, representative data from 1 of 3 patients are shown. (D and E) NETs were identified in AAA specimens from patients with EH (confocal microscopy: blue, DAPI; green, NE; red, CitH3; original magnification, 400×), bearing TF as indicated in (E) (confocal microscopy: blue, DAPI; green, TF; red, NE; original magnification, 630×). For D and E, representative data from 2 of 3 patients are shown. Yellow asterisk indicates the luminal site of the AAA, and white asterisk indicates disrupted elastic lamina.