Myeloperoxidase impacts vascular function by altering perivascular adipocytes' secretome and phenotype in obesity

Abstract

Obesity, a main driver of cardiovascular morbidity, contributes to endothelial dysfunction and inflammation in adipose tissues. Perivascular adipose tissue (PVAT) surrounds arteries and influences vascular function. In obesity, immune cells, including myeloperoxidase (MPO)-releasing myeloid cells, accumulate in PVAT. In this study, we show MPO levels to correlate with body weight and endothelial function in obese patients (n = 33) and mice. In addition, MPO deficiency reduces immune cell frequency, enhances PVAT beiging via soluble guanylyl cyclase β1 (sGC-β1), and increases oxygen consumption in vivo. Further, nitrotyrosine formation and inflammatory cytokine release are attenuated in obese Mpo^-/- mice. Mechanistically, adiponectin (APN) secretion improves endothelial function and reduces arterial stiffness. In vitro, MPO-treated human white adipocytes show lower APN and brown adipocyte marker expression but increased inflammation. Thus, MPO impairs vascular function via PVAT inflammation and suppression of vasoprotective mediators, making it a potential therapeutic target in obesity-related cardiovascular disease.

Keywords: adiponectin; endothelial function; inflammation; myeloperoxidase; obesity; perivascular adipose tissue.

Graphical abstract

Figure 1

Bodyweight loss reduces inflammatory state and improves endothelial function in bariatric patients Schematic overview of the experimental setup (A). Obese patients with a BMI > 37 kg/m² underwent bariatric surgery by Roux-en-Y gastric bypass or gastric banding. Blood samples were obtained, and flow-mediated dilation (FMD) was measured before (pre) and 3 months after (post) bariatric surgery. Lean patients were included for blood analyses as non-obese controls. BMI (B), blood cholesterol levels (C), and plasma high-sensitive CRP (hsCRP) (D) before and after bariatric surgery and in lean control patients. MPO plasma levels (E) and correlation of BMI and MPO plasma levels (F). FMD analysis before and after bariatric surgery (G, green area indicates FMD standard value range¹⁸) and correlation of FMD and MPO plasma levels (H). n (patients) = 33. n (lean controls) = 14. For (E) and (G), pre- and post-OP values were analyzed. p as indicated. Statistical significance was determined by ordinary one-way ANOVA followed by Tukey’s multiple comparison test (C, E, and G) and Kruskal-Wallis followed by Dunn’s multiple comparison test (B and D) or simple linear regression using a Pearson’s correlation calculation (F and H).

Figure 2

MPO impacts vascular function in murine models of obesity Schematic overview of the experimental setup (A). For dietary-induced obesity (DIO), wild-type (WT) and Mpo^−/− mice were fed a 60% HFD for 12 weeks. Leptin (Lep)^−/− mice with genetically inherited obesity (GIO) were crossbred with Mpo^−/− mice. Lean WT and Mpo^−/− littermates of the respective line were used as controls. Body weight curve of HFD-fed WT and Mpo^−/− mice and lean controls over 12 weeks starting at 8 weeks of age; n = 4 (B). Body weight of WT, Lep^−/−, Lep^−/−/Mpo^−/−, and Mpo^−/− mice at 14 weeks of age (C). MPO levels in plasma (D) and in perivascular adipose tissue (PVAT) (E) of obese mice and lean controls in the DIO (left) and GIO (right) model. Acetylcholine-depended endothelium-mediated vascular relaxation of isolated aortic rings of DIO (F) and GIO mice (G) and lean controls measured by organ bath investigation. Representative images of vascular ultrasound of carotid arteries (H). Pulse wave velocity (PWV) analyses in obese mice and lean controls in the DIO (I) or GIO (J) model. Data are presented as mean ± SEM. n = as indicated. Statistical significance was determined by unpaired Student’s t test (D and E) or by ordinary two-way ANOVA followed by Tukey’s multiple comparison test (B, C, F, G, I, and J). (B, F, and G) p = WT HFD vs. Mpo^−/−HFD or Lep^−/− vs. Lep^−/−/Mpo^−/−, respectively. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. n = 5–13

Figure 3

MPO deficiency reduces myeloid cell frequency in perivascular adipose tissue after high-fat diet Leukocyte populations in DIO PVAT were analyzed by flow cytometry (A: gating strategy). Counts of CD45⁺ leukocytes (B), CD45⁺CD11b⁺ myeloid cells (C), CD45⁺CD11b⁺CD64⁺Ly6G⁺ neutrophils (D), CD45⁺CD11b⁺CD64⁺Ly6c⁺ monocytes (E), and CD45⁺CD11b⁺CD64⁺F4/80⁺ macrophages (F) are shown. Representative immunohistochemical stainings of Ly6G⁺ neutrophils in PVAT of DIO mice (G). White arrows indicate Ly6G⁺ neutrophils. Ly6G count in the PVAT of DIO (H) and GIO (I) mice. n = as indicated. Data are presented as mean ± SEM. Statistical significance was determined by ordinary two-way ANOVA followed by Tukey’s multiple comparison test.

Figure 4

Myeloperoxidase mediates nitrosative tissue damage and pro-inflammatory signaling of PVAT in obesity Gene expression of IL-1β, TNF-α, and CCL-2 in PVAT of DIO mice (A–C). Representative immunohistochemical stainings of nitrosative tissue damage in the PVAT of obese WT and Mpo^−/− mice and in lean controls (D). Evaluation of nitrotyrosine in the PVAT of DIO (E) and GIO (F) mice. Data are presented as mean ± SEM. Statistical significance was determined by ordinary two-way ANOVA followed by Tukey’s multiple comparison test. n = as indicated.

Figure 5

MPO impedes PVAT beiging by inhibition of soluble guanylyl cyclase β1 PVAT morphology of DIO WT and Mpo^−/− was assessed by automated quantification of lipid vacuoles in hematoxylin and eosin stainings (A: representative H&E stainings and automated binarization; B: results of lipid droplet quantification). UCP-1 mRNA (C and D), UCP-1 protein (E and F), and sGC-β1 protein expression in DIO and GIO PVAT (G and H). Representative immunoblots of UCP-1, sGC-β1, and GAPDH from DIO (I) and GIO (J) PVAT. GAPDH was used as loading control. Data are presented as mean ± SEM. Statistical significance was determined by ordinary two-way ANOVA followed by Tukey’s multiple comparison test. n = as indicated. UCP-1, uncoupling protein 1; sGC-β1, soluble guanylyl cyclase β1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Figure 6

MPO impairs endothelial function by impacting paracrine activity of perivascular adipose tissue APN mRNA and protein expression and representative immunoblots of APN and GAPDH of DIO (A–C) and GIO (D–F) PVAT. mRNA expression of PPARγ, PGC-1α, AMPK, and eNOS in the PVAT of GIO mice (G–J). GAPDH was used as loading control. PVAT was cultivated for 12 h in DMEM. Isolated aortic rings were incubated with the indicated PVAT supernatants, and organ bath experiments were performed (K). Acetylcholine-depended endothelium-mediated vascular relaxation of isolated aortic rings of Adipoq^−/− mice, aortic rings of Mpo^−/− mice incubated with Adipoq^−/− PVAT (Mpo^−/− PVAT^{Adipoq−/−}) vice versa, and respective controls (L). MPO protein levels of phorbol 12-myristate 13-acetate (PMA)- and APN-treated PMN-like HL-60 cells within the supernatant (M) and in the cells (N). Data are presented as mean ± SEM. Statistical significance was determined by ordinary two-way ANOVA (A–D, G–J, and L) or by ordinary one-way ANOVA followed by Tukey’s multiple comparison test (M and N). For (L), significances are shown for Mpo^−/−PVAT^{Adipoq−/−} vs. Adipoq^−/−PVAT^Mpo−/−: ∗p < 0.05. n = as indicated. APN, adiponectin; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; PPARγ, peroxisome proliferator-activated receptor gamma; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; AMPK, AMP-activated protein kinase; eNOS, endothelial nitric oxide synthase; Mpo^−/−, MPO-deficient; Adipoq^−/−, adiponectin-deficient.

Figure 7

MPO changes adipocyte phenotype, cytokine, and adipokine expression in vitro Subcutaneous human white preadipocytes (sHWPs) were differentiated to adipocytes (sHWAs), identified by excessive lipid vacuole formation (A, red circles in oil red O staining). Expression of ASC-1, UCP-1, CITED-1, and APN mRNA of sHWA incubated at 32°C and stimulated with noradrenaline (Nor) for BAT induction and control cells incubated at 37°C without Nor treatment (B–E). UCP-1, CITED-1, ASC-1, IL-1β, IL-6, and APN mRNA expression of MPO/H₂O₂-treated or control-treated beiged sHWA (F–K). APN protein expression of MPO/H₂O₂- or control-treated sHWA (L and M). α-tubulin was used as loading control. Data are presented as mean ± SEM. Statistical significance was determined by unpaired Student’s t test; for (C), a non-parametric Mann-Whitney test was used. n = as indicated. ASC-1, Asc-type amino acid transporter 1; UCP-1, uncoupling protein 1; CITED-1, Cbp/p300-interacting transactivator 1; APN, adiponectin; IL-1β, interleukin 1β; IL-6, interleukin 6; BAT, brown adipose tissue.