Immunoproteasomal Processing of IsoLG-Adducted Proteins Is Essential for Hypertension

Abstract

Background: Hypertension is characterized by CD8⁺ (cluster differentiation 8) T cell activation and infiltration into peripheral tissues. CD8⁺ T cell activation requires proteasomal processing of antigenic proteins. It has become clear that isoLG (isolevuglandin)-adduced peptides are antigenic in hypertension; however, IsoLGs inhibit the constitutive proteasome. We hypothesized that immunoproteasomal processing of isoLG-adducts is essential for CD8⁺ T cell activation and inflammation in hypertension.

Methods: IsoLG adduct processing was studied in murine dendritic cells (DCs), endothelial cells (ECs), and B8 fibroblasts. The role of the proteasome and the immunoproteasome in Ang II (angiotensin II)-induced hypertension was studied in C57BL/6 mice treated with bortezomib or the immunoproteasome inhibitor PR-957 and by studying mice lacking 3 critical immunoproteasome subunits (triple knockout mouse). We also examined hypertension in mice lacking the critical immunoproteasome subunit LMP7 (large multifunctional peptidase 7) specifically in either DCs or ECs.

Results: We found that oxidant stress increases the presence of isoLG adducts within MHC-I (class I major histocompatibility complex), and immunoproteasome overexpression augments this. Pharmacological or genetic inhibition of the immunoproteasome attenuated hypertension and tissue inflammation. Conditional deletion of LMP7 in either DCs or ECs attenuated hypertension and vascular inflammation. Finally, we defined the role of the innate immune receptors STING (stimulator of interferon genes) and TLR7/8 (toll-like receptor 7/8) as drivers of LMP7 expression in ECs.

Conclusions: These studies define a previously unknown role of the immunoproteasome in DCs and ECs in CD8⁺ T cell activation. The immunoproteasome in DCs and ECs is critical for isoLG-adduct presentation to CD8⁺ T cells, and in the endothelium, this guides homing and infiltration of T cells to specific tissues.

Keywords: LMP7 protein; dendritic cells; endothelial cells; hypertension; isolevuglandin.

Figure 1:. IsoLG-adduct interaction with H-2D^b is Dependent on Proteasome Activity.

Intermolecular interaction of IsoLG-adducts with the MHC-I (class I major histocompatability complex) H-2D^b was determined by FRET. (A) Paradigm of detection of isoLG-adduct H-2D^b interaction by FRET. A donor fluorophore was conjugated to an anti MHC-I antibody and an acceptor fluorophore was conjugated to the D11 ScFv anti-isoLG adduct antibody that recognizes the IsoLG adduct (IA). The donor fluorophore was excited by a laser at energy hv and the emission of the donor excites the acceptor fluorophore which emits photons at energy hv’ that is detected by flow cytometry. (B) Representative histograms of surface isoLG-adducts from DCs treated with tBHP and proteasome inhibitors MG132 and BTZ. (C) Quantitation of isoLG positive DCs. (D) Representative histograms of FRET from DCs treated with tBHP. (E) Quantitation of FRET⁺ DCs. Data are representative of 3 repeated experiments. Data were analyzed by 1-way ANOVA with Tukey’s post-hoc test (N = 3)

Figure 2:. Inhibition of LMP7 attenuates hypertension and aortic T cell infiltration.

Mice were co-treated with angiotensin II and vehicle or PR-957 for 2-weeks. (A) Systolic blood pressure was measured. Data were analyzed by two-tailed Student’s T-test (8 animals per group). (B) Representative flow cytometry plots of isoLG-adduct-H-2D^b FRET in dendritic cells harvested from aorta (4 animals per group). (C) Quantitation of isoLG-adduct-H-2D^b FRET⁺ dendritic cells harvested from aorta. (D) Representative flow cytometry plots of CD3⁺ T cells harvested from aorta. (E) Quantitation of CD3⁺ (F) CD8⁺ and CD4⁺ T cells harvested from aorta. Data were analyzed by 1-way ANOVA with Tukey’s post-hoc test (sham: 4 animals, Ang II: 8 animals, and AngII+PR957: 8 animals).

Figure 3:. Overexpression of immunoproteasome subunits increases isoLG-adduct MHC-I interaction.

B8 and B8.27M.2 cells were treated with vehicle or tBHP. (A) Representative flow cytometry plots of FRET. (B) Representative histograms of FRET. (C) Quantitation of percent of FRET⁺ cells as a percentage of H2-D^b+/IsoLG-adduct⁺ cells. (D) Quantitation of FRET⁺ cells. Data were analyzed using 1-way ANOVA with Tukey’s post-hoc test (B8, B8 + TBHP, and B8.27M.2 + TBHP: n = 6; B8.27M.2: n = 5).

Figure 4:. The immunoproteasome contributes to the development of hypertension.

Immunoproteasome triple knockout mice (TKO) and control C57BL/6 mice were treated with angiotensin II. Radiotelemetry was used to measure (A) systolic, (B) diastolic, and (C) mean blood pressure. Blood pressure was analyzed by mixed-effects analysis (WT AngII: 7 animals; TKO AngII: 5 animals). (D) Representative flow cytometry diagrams for CD4⁺ and CD8⁺ T cells harvested from kidney. (E) Quantitation of CD8⁺ renal T cells. (F) Representation of CD8⁺ T cells as a percentage of total CD3⁺ T cells. (G) Quantitation of total CD3⁺ renal T cells (WT AngII: 7 animals; TKO AngII 5 animals). (H) Quantitation of FRET⁺ DCs harvested from aorta (6 mice per group). (I) Representation of FRET⁺ aortic DCs as a percentage of total DCs. (J) Representative flow cytometry plots for FRET in DCs harvested from aorta. Quantitation of (K) total CD3⁺ and (L) CD8⁺ T cells harvested from aorta (4 animals per group). Data were analyzed by two-tailed Student’s T-test or Mann Whitney U Test.

Figure 5:. DC-specific LMP7 expression contributes to hypertension:

Radiotelemetry was used to measure (A) systolic (B) diastolic (C) and mean arterial pressure (MAP). Blood pressure was analyzed using mixed-effects analysis (Psmb8^fl/fl: 6 animals, Psmb8^fl/fl/CD11c-Cre: 5 animals). (D) Representative histogram of isoLG-adduct MHC-I FRET in DCs from angiotensin II treated Psmb8^fl/fl and Psmb8^fl/fl/CD11c-Cre mice. (E) Quantitation of FRET⁺ in DCs, macrophages (MF), Ly6c high monocytes, Ly6c intermediate monocytes, and L76c low monocytes harvested from aorta. Data were analyzed by by 2-way ANOVA with Šídák's multiple comparison test (Psmb8^fl/fl : 7 animals, Psmb8^fl/fl/CD11c-Cre: 6 animals). (F) Quantitation of CD3⁺ T cells harvested from aorta (Psmb8^fl/fl: 7 animals, Psmb8^fl/fl/CD11c-Cre: 6 animals). Data were analyzed by two-tailed Student’s T-test or Mann-Whitney U Test.

Figure 6.. ROS Increases IsoLG-Adduct-H-2D^b interaction in ECs and drives T cell proliferation.

Mouse aortic endothelial cells were treated with glucose oxidase and flow cytometry was used to quantitate (A) ICAM-1, (B) I-Ab, (C) IsoLG, and (D) IsoLG-Adduct-H-2D^b FRET (N = 6). Following infusion of sham or Ang II aortic ECs were evaluated for IsoLG-Adduct-H-2D^b FRET (E) Quantitation of FRET⁺ ECs. Data were analyzed by a two-tailed Student’s T-test or Mann-Whitney U Test (7 animals per group) (F) Representative histogram of FRET in aortic ECs. CD8⁺ T cells isolated from sham or AngII treated mice were co-cultured with GO or control treated mouse aortic endothelial cells and evaluated for proliferation. (G) Quantitation of proliferation of CD8⁺ T cells. Data were analyzed by a one-way ANOVA with Tukey’s post-hoc test (n = 6). (H) Representative histogram of CFSE staining of CD8⁺ T cells following co-culture.

Figure 7:. EC-specific LMP7 expression contributes to hypertension and aortic inflammation.

(A) Immunoblot for LMP7 and β-Actin on protein isolated from ECs sorted from the aorta of Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice. (B) Immunoblot for LMP7 and β-Actin on protein isolated from spleen, kidney, and heart of Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice. Each lane represents an individual mouse (N = 2). (C) Radiotelemetry tracings of systolic blood pressure from Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice treated with Ang II. Data was analyzed by 2-way ANOVA (5 mice per group) (D) Quantitation and (E) representative histogram of FRET⁺ ECs from aorta of Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice treated with Ang II. Data were analyzed by two-tailed student’s T test (Psmb8^fl/fl : 6 animals; Psmb8^fl/fl/VECAD-Cre: 7 animals). (F) Quantitation and (G) representative flow plots of infiltrating CD8⁺ T cells from aorta of Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice treated with Ang II. (H) Quantitation and (I) representative flow plots of infiltrating DCs from aorta of Psmb8^fl/fl and Psmb8^fl/fl/VECAD-Cre mice treated with Ang II (Psmb8^fl/fl : 8 animals; Psmb8^fl/fl/VECAD-Cre: 9 animals). Data was analyzed by two-tailed Student’s T test.

Figure 8:. IsoLGs regulate inflammatory gene expression in ECs and contribute to STING activation.

(A) Comparison of gene program activation in the ethyl-2-hydroxybenzylamine (Et2HOBA) + glucose oxidase (GO) treated mouse aortic endothelial cells (MAECs) group against the GO group was carried out through a normalized enrichment score with the R-based program Web Gestalt. Random sampling of equally sized gene sets is conducted to determine significance and to normalize the data. Values were expressed through the false discovery rate (FDR). (B) Immunoblot analysis for β5, LMP7, and H2-D^b with β-actin loading controls from MAECs treated with control, control+Et2HOBA, GO, or GO+Et2HOBA (n=2). (C) Chymotrypsin proteasome activity in ECs following treatment with control or GO. Data was analyzed by a two-tailed Student’s T-test (N=7). (D) Sybr green and Coomassie blue staining of isoLG adducted BSA at increasing concentrations (E) Immunoblot for phospho-STING^ser365 (pSTING) and total STING in MAECS treated with control, control+Et2HOBA, GO, or GO+Et2HOBA (n=2). (F) Immunoblot for phospho-IRAK4 and total IRAK4 in MAECs treated GO and GO+Et2HOBA (n=4). (G) Immunoblot for LMP7 and β-actin in MAECs treated with tert-butyl-hydroperoxide (TBHP) and TBHP + the STING inhibitor C176 (n=3) (H) Immunoblot for LMP7 and β-actin in MAECs treated with TBHP and TBHP + the TLR7/8 inhibitor Enpatoran (n=3).