Myeloid-Derived Suppressor Cells Restrain Natural Killer Cell Activity in Acute Coxsackievirus B3-Induced Myocarditis

Figures

• 

  Figure 1. Granulocytic and monocytic MDSC subpopulations are increased in the spleen and heart of CVB3-infected A.BY/SnJ mice. Isolated splenic and cardiac immune cells were selected based on CD11b+Ly6GhighLy6Clow (granulocytic MDSC; (A,C)) and on CD11b+Ly6GlowLy6Chigh (monocytic MDSC; (B,D)). The y-axis displays the percentual proportion of 50,000 and 20,000 measured splenic or cardiac cells, respectively. The panel below (E) shows representative immunofluorescence pictures of cardiac tissue 4 days post infection (dpi) stained with anti-CD11b-AF488, anti-Gr1-AF647 (monocytic), anti-Ly6G-AF647 (granulocytic), and DAPI. Representative images of flow cytometry of the spleen from naive mice (upper panel) and from mice 0 dpi, 4 dpi, 8 dpi CVB3 (lower panel) are shown in (F). The gating strategy is depicted as R1 = vital cells, R2 = CD11b+ cells, R3 = Ly6Ghigh granulocytic MDSC, R4 = monocytic MDSC. Data are expressed as mean ± SEM and were analyzed with One-way ANOVA or Kruskal-Wallis test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 with n = 13 for 0 dpi, n = 8 for 4 dpi, and n = 6 for 8 dpi for the splenic immune cells (A,B) and n = 4/group for the cardiac immune cells (C,D)).

• 

  Figure 2. MDSC increased the expression of CD107a, granzyme B, and NKG2D in NK cells in vitro. Impact of Ly6G MDSC (granulocytic) or Ly6C MDSC (monocytic) on NK cells (NKp46+) in presence of CVB3-infected co-cultured HeLa/RAW cells. (A) Effect on the degranulation marker CD107a (B) granzyme B expression and (C) NKG2D receptor expression on NKp46+ NK cells. The y-axis displays the percentual proportion of 20,000 measured splenic NK cells. Data are expressed as mean ± SEM and were analyzed with One-way ANOVA or Kruskal-Wallis test (** p < 0.01, *** p < 0.001, **** p < 0.0001 with n = 3–6 for NKp46+, n = 3 for NKp46++Ly6G MDSC, and n = 3 for NKp46++Ly6C MDSC).

• 

  Figure 3. Gr-1 depletion decreases viral load, damage, and inflammation in the heart of CVB3-infected A.BY/SnJ mice. (A) Following treatment of CVB3-infected mice with anti-Ly6G (RB6-8C5), spleen cells were stained at 0 dpi and 4 dpi with an anti-Gr-1-FITC antibody and analyzed via flow cytometry. The y-axis displays the percentual proportion of 20,000 measured splenic cells (with n = 4–7 for 0 dpi and n = 4–9 for 4 dpi). (B) Next, hearts from Gr-1-depleted and non-depleted mice were analyzed regarding CVB3 copy number by quantitative real-time PCR at 0 dpi, 4 dpi, and 8 dpi (with n = 5 for 0 dpi, n = 5 for 4 dpi, and n = 4–5 for 8 dpi). (C) In addition, myocardial damage was quantitatively determined on hematoxylin-eosin (H/E) stained tissue sections at the respective time points (with n = 3–4 for 0 dpi, n = 4 for 4 dpi, and n = 5 for 8 dpi). Additionally, Gr-1 depletion decreases Mac3+ macrophages and CD3+ T lymphocytes in the heart of CVB3-infected A.BY/SnJ mice. After depletion of Gr-1+ cells with anti-Ly6G (RB6-8C5), the number of (D) Mac3+ macrophages and (E) CD3+ T cells/ mm2 heart area (HA) was quantified at 0, 4, and 8 dpi, using the respective immunohistological stainings. (F)…

• 

  Figure 4. S100A8 and S100A9 mRNA and protein expression are increased in CVB3-infected A.BY/SnJ mice. mRNA levels of cardiac (A) S100A8, and (B) S100A9, and splenic (C) S100A8 and (D) S100A9 were measured by quantitative real-time PCR at 0, 4, and 8 dpi (with n = 5 for 0 dpi, n = 4 for 4 dpi, and n = 4 for 8 dpi) for the cardiac gene expression levels (A,B) and n = 4/group for the splenic gene expression levels (C,D). (E) S100A8 and (F) S100A9 protein expression in splenic CD11b+Gr-1+ cells measured by flow cytometry (n = 5). The y-axis displays the percentual proportion of 20,000 measured cardiac and splenic cells. Data are expressed as mean ± SEM and were analyzed with One-way ANOVA or Kruskal-Wallis test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

• 

  Figure 5. Cardiac S100A8- and S100A9-expressing cells are present before CD3+ T lymphocyte infiltration, and Gr-1 depletion results in a decrease of S100A8 protein levels in the heart of CVB3-infected A.BY/SnJ mice. (A) Representative H/E and immunohistological stainings for CD3+ T lymphocytes, S100A8, and S100A9 (see circles for positive green cells) in serial heart tissue sections of CVB3-infected mice 4 dpi (magnification ×400). Hearts from CVB3-infected mice with or without Gr-1-depletion were subjected to (B) H/E (upper panel), and immunohistochemical stainings of Mac3+ macrophages (middle panel, green cells) and S100A8 protein (lower panel, green cells) 8 dpi (magnification x400). (C) Subsequent quantification of S100A8 protein expression as number of positive cells/mm2 HA. Data are expressed as mean ± SEM and were analyzed with One-way ANOVA (**** p < 0.0001 with n = 5 for 0 dpi, n = 5 for 4 dpi, and n = 5 for 8 dpi).

• 

  Figure 6. Gr-1 depletion decreases cardiac IL-1β, IL-6, and TNF-α mRNA levels in CVB3-infected A.BY/SnJ mice. Hearts of Gr-1-depleted and non-depleted CVB3-infected mice were subjected to (A) IL-1β, (B) IL-6, and (C) TNF-α mRNA quantification by real-time RT-PCR at 0, 4, and 8 dpi. Furthermore, Gr-1 depletion decreases TNF-α, and MIP-1α mRNA levels in splenic NK cells of CVB3-infected A.BY/SnJ mice. In detail, mRNA levels were quantified in splenic NK cells 0, 4, and 8 dpi of (D) TNF-α and (E) MIP-1α after Gr-1 depletion. Data are expressed as mean ± SEM and were analyzed with One-way ANOVA or Kruskal-Wallis test (** p < 0.01, *** p < 0.001, **** p < 0.0001 with n = 5 for 0 dpi, n = 3–4 for 4 dpi, and n = 4–5 for 8 dpi).