Immunoreceptor CD300a regulates ischemic tissue damage and adverse remodeling in the mouse heart and kidney

Abstract

Acute ischemic organ diseases such as acute myocardial infarction and acute kidney injury often result in irreversible tissue damage and progress to chronic heart failure (CHF) and chronic kidney disease (CKD), respectively. However, the molecular mechanisms underlying the development of CHF and CKD remain incompletely understood. Here, we show that mice deficient in CD300a, an inhibitory immunoreceptor expressed on myeloid cells, showed enhanced efferocytosis by tissue-resident macrophages and decreased damage-associated molecular patterns and pathogenic SiglecFhi neutrophils, resulting in milder inflammation-associated tissue injury than in wild-type mice after ischemia and reperfusion (IR). Notably, we uncovered that CD300a deficiency on SiglecFlo neutrophils increased the signal transducer and activator of transcription 3-mediated production of pro-angiogenic and antifibrotic factors, resulting in milder adverse remodeling after IR. Our results demonstrated that CD300a plays an important role in the pathogenesis of ischemic tissue injury and adverse remodeling in the heart and kidney.

Keywords: Fibrosis; Immunology; Inflammation; Macrophages; Neutrophils.

Figure 1. CD300a deletion ameliorates myocardial ischemia and reperfusion injury and adverse remodeling.

(A) MI/R model. LAD, left anterior descending artery. (B) Plasma cTnI in Cd300a^–/– (n = 11) and WT mice (n = 15). (C) Evans blue and TTC staining (left) and the percentage of infarct area among ischemic area in Cd300a^–/– (n = 6) and WT mice (n = 9). (D) Representative immunohistochemistry (left) and quantitative data of CD31⁺ capillary density in peri-infarct area in Cd300a^–/– (n = 4 [day 0], 8 [day 3], and 9 [day 14]) and WT mice (n = 5 [day 0], 6 [day 3], and 8 [day 14]). Scale bars: 100 μm. (E) Representative immunohistochemistry (left) and the percentage of FSP1⁺ and αSMA⁺ fibroblasts in vimentin⁺ fibroblasts in infarct area in Cd300a^–/– (n = 8) and WT mice (n = 7). Scale bars: 100 μm. (F) Top: Representative Masson’s trichrome staining of the heart. Bottom: Fibrosis area in Cd300a^–/– (n = 12 in MI/R and n = 2 in sham) and WT heart (n = 15 in MI/R and n = 2 in sham). Scale bar: 1 mm. (G) Representative echocardiography images (left) and left ventricular internal dimension in systole (LVDs), fractional shortening (LVFS), and ejection fraction (LVEF) (n = 2 in each group) in Cd300a^–/– mice (n = 12) and WT mice (n = 15). Scale bars: 2 mm. Data are presented as mean ± SEM and pooled from more than 5 (B, C, F, and G) and 3 experiments (D and E). Dots represent independent animals. Statistical analysis was performed using unpaired Student’s t test (B, C, and E) and 2-way ANOVA (D, F, and G). *P < 0.05; **P < 0.01; ****P < 0.0001.

Figure 2. CD300a deficiency ameliorates AKI and fibrosis after biIRI.

(A) Bilateral ischemia-reperfusion injury (biIRI) model. (B) Plasma NGAL in sham operation (n = 4 in each group) and after biIRI (Cd300a^fl/fl mice, n = 3 [0 hours], 4 [6 hours], 13 [day 1], 7 [day 2], 6 [day 3], and 4 [day 5]; Cd300a^fl/flLyz2-Cre mice, n = 3 [0 hours], 4 [6 hours], 13 [day 1], 6 [day 2], 4 [day 3], and 4 [day 5]). (C) Plasma BUN and Cre in sham (n = 4) and after biIRI (n = 5 [day 0], 31 [day 1], 24 [day 2], 9 [day 3], and 9 [day 5]) in each group. (D) Representative periodic acid–Schiff staining of kidneys after biIRI in Cd300a^fl/fl and Cd300a^fl/flLyz2-Cre mice (n = 6 or 7). Arrows, loss of brush border; arrowheads, necrosis; stars, debris. (E) Left: Representative KIM-1 and p-H3 staining of kidneys after biIRI. Right: The percentage of p-H3⁺ cells in KIM-1⁺ cells of Cd300a^fl/fl (n = 2 [0], 6 [2 days], and 5 [7 days]) and Cd300a^fl/fl Lyz2-Cre mice (n = 2, 6, and 4). (F) Tgfb, Ctgf, and Il1b mRNA expression in the kidneys in sham (n = 2) and after biIRI (n = 3 [days 0, 1, and 5] and 5 [day 7] in each group). Relative quantity (RQ). (G) Left: Representative Sirius red staining of kidneys after biIRI. Right: Fibrosis in tubulointerstitial area in Cd300a^fl/fl and Cd300a^fl/flLyz2-Cre mice (n = 7 or 5). Scale bars: 100 μm (D, E, and G). Data are presented as mean ± SEM and pooled of 2 (B, D, F, and G), 3 (E), and 5 (C) experiments. Statistical analyses were performed using unpaired Student’s t test (B–D, F, and G) and 2-way ANOVA (E). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 3. Enhanced efferocytosis reduces tissue damage and fibrosis and preserves organ function after MI/R and biIRI.

(A and B) Efferocytosis of PSVue⁺ cells by CD11b⁺ myeloid cells in the heart of Cd300a^–/– (n = 10) and WT mice (n = 9). (C) Plasma HMGB-1 and IL-1α in Cd300a^–/– and WT mice (n = 23 or 24 in each group). (D and E) Efferocytosis of host-derived EGFP⁺ dead cells by donor-derived tdTomato⁺CD11b⁺ myeloid cells in the tdTomato-GFP chimeric mouse kidneys (n = 3 or 4 in each group). (F) Plasma HMGB-1 in Cd300a^fl/fl and Cd300a^fl/flLyz2-Cre mice (n = 3 [0 hours] and 7 or 8 [24 hours]). (G and H) Plasma cTnI and LVEF in control EPT (Ctrl)- or D89E-administered Cd300a^–/– mice (Ctrl, n = 7 or 6; D89E, n = 8 or 4) and WT mice (Ctrl, n = 5 or 4; D89E, n = 7 or 6). (I and J) Plasma NGAL and BUN and Cre in Ctrl- or D89E-administered Cd300a^fl/fl (n = 8 or 5 for I and n = 15 or 14 for J) and Cd300a^fl/flLyz2-Cre mice (n = 6 or 5 for I and n = 11 or 10 for J). (K and L) Tgfb expression and the fibrosis area by Masson’s trichrome staining in the kidneys of Ctrl- or D89E-administered Cd300a^fl/fl (n = 5 in each group for K and n = 5 or 6 for L) and Cd300a^fl/flLyz2-Cre mice (n = 5 or 4 for K and n = 5 in each group for L). Data are presented as mean ± SEM and pooled of 2 (E, I, K, and L), 3 (B, C, and F–H), and 5 (J) experiments. Two-way ANOVA (B and E–L) and unpaired Student’s t test (C). *P < 0.05; **P < 0.01.

Figure 4. CD300a deficiency upregulates pro-angiogenic and antifibrotic genes in SiglecF^lo neutrophils.

(A–D) CD31⁺ capillary density in peri-infarct area (A), the percentage of FSP1⁺ cells (B), LVFS and LVEF (C), and the percentage of scar areas (D) in isotype control (Ctrl)– or anti-Ly6G mAb–administered Cd300a^–/– and WT mice (Ctrl, n = 6–8 [A and B], 10–12 [C], and 13–16 [D]). Scale bars: 1 mm. (E) CD31⁺ capillary density in the peri-infarct area in the heart of mice into which WT or Cd300a^–/– neutrophils were transferred. Scale bars: 100 μm. (F) Flow cytometric analysis of SiglecF^hi and SiglecF^lo neutrophils in the cardiac tissue in WT (n = 12) and Cd300a^–/– (n = 11) mice. (G) The correlation between LVEF and the percentages of SiglecF^lo neutrophils. (H) Principal component analysis plot of RNA expression of SiglecF^hi and SiglecF^lo neutrophils in the cardiac tissue from WT and Cd300a^–/– mice (n = 8 in each group). (I) GSEA profiles of enrichment gene sets associated with angiogenesis in Cd300a^–/– and WT SiglecF^lo neutrophils. (J) Heatmap of enriched genes associated with angiogenesis. (K) Prok2 and Chil1 mRNA expression in SiglecF^hi and SiglecF^lo neutrophils in cardiac tissue in Cd300a^–/– and WT mice (n = 10 or 7). (L) Heatmap of enriched genes associated with degradation of the extracellular matrix. (M) Mmp9 mRNA expression in SiglecF^lo neutrophils in the cardiac tissue in Cd300a^–/– and WT mice (n = 7 or 10). Data are presented as mean ± SEM and pooled from 2 (E–G) and more than 5 (A–D, K, and M) experiments. Two-way ANOVA (A–D and K) and unpaired Student’s t test (E, F, and M). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 5. CD300a suppresses STAT3 phosphorylation in SiglecF^lo neutrophils.

(A–C) WT or Cd300a^−/− BM neutrophils were cultured for 2 days in the presence of the culture supernatant of the naive cardiac tissue, which had been stimulated or not with HMGB-1 or IL-1α (A) and analyzed for SiglecF expression by flow cytometry (B). (C) The GM-CSF and G-CSF levels in the culture supernatant of the cardiac tissue after stimulation with HMGB-1 or IL-1α. (D) Flow cytometric analysis of SiglecF expression on WT or Cd300a^−/− BM neutrophils after stimulation with GM-CSF or G-CSF. (E) Ingenuity Pathway Analysis based on differentially expressed genes from Cd300a^–/– and WT SiglecF^lo neutrophils. (F and G) Flow cytometric analysis of phosphorylated STAT3 (p-STAT3) in neutrophils of the cardiac tissue of Cd300a^–/– (n = 7) and WT (n = 6) mice (F) and WT or Cd300a^–/– BM neutrophils stimulated with or without the cardiac tissue supernatant (G). (H and I) Prok2 and Chil1 mRNA expression in BM neutrophils of WT or Cd300a^–/– mice stimulated with the cardiac tissue supernatant after MI/R (H) or with G-CSF together with STAT3 inhibitor (STAT3i) or control vehicle (Vehicle) (I). (J) LVEF after MI/R in Cd300a^–/– (vehicle, n = 5; STAT3i, n = 4) and WT mice (n = 3 in each group) treated with STAT3i or vehicle. Data are presented as mean ± SEM, representative of 2 experiments (B, D, G, and H) and pooled from 2 (C, F, and I) and 4 (J) experiments. One-way (B–D) and 2-way ANOVA (F, G, I, and J) and unpaired Student’s t test (H). *P < 0.05; **P < 0.01; ****P < 0.0001.

Figure 6. Anti-CD300a mAb ameliorates tissue injury and adverse remodeling.

(A) Plasma cTnI in anti-CD300a mAb– or Ctrl-administered mice (n = 7 or 8). (B) CD31⁺ capillary density in peri-infarct area in anti-CD300a mAb– or Ctrl-administered mice (n = 7). (C) LVFS and LVEF in anti-CD300a mAb– or Ctrl-administered mice (n = 7 or 5). (D) Masson’s trichrome staining and the fibrosis area in the ventricle in anti-CD300a mAb– or Ctrl-administered mice (n = 10 or 11). (E) Efferocytosis of GFP⁺ dead cells by myeloid cell subpopulations in the cardiac tissue of αMHC-GFP mice given anti-CD300a mAb or Ctrl (n = 6 or 7). (F) Generation of BM chimeric R26GRR mice (GFP⁺ mice) and efferocytosis of host-derived GFP⁺ dead cells by resident macrophages of the chimeric mice given an anti-CD300a mAb (n = 8 or 5) or Ctrl (n = 5 or 4). (G) Plasma BUN and Cre in WT and Cd300b^–/– mice given anti-CD300a mAb or Ctrl (WT, n = 3 or 17; Cd300b^–/–, n = 5 or 6). (H) Periodic acid–Schiff staining of the kidney and acute tubular necrosis (ATN) score in mice given anti-CD300a mAb or Ctrl (n = 9 in each group). (I) Sirius red staining and the fibrosis area in WT and Cd300b^–/– mice given anti-CD300a mAb or Ctrl (WT, n = 8; Cd300b^–/–, n = 5 or 6). Scale bars: 100 μm (B, H, and I) and 1 mm (D). Data are presented as mean ± SEM and pooled from 2 (A–D, F, H, and I) and 5 (E and G) experiments. Student’s t test (A–D and H) and 2-way ANOVA (E–G and I). *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 7. Blockade of human CD300A enhances efferocytosis and ameliorates acute renal injury after biIRI.

(A and B) A public gene expression dataset of patients with CKD of diabetic nephropathy (DN) (n = 9) and focal segmental glomerulosclerosis (FSGS) (n = 19). CD300A expression in comparison with the healthy control group (n = 10) (A) and Spearman’s correlation analysis between the expression of CD300A and renal function of glomerular filtration rate (GFR) and creatinine (B). MDRD, Modification of Diet in Renal Disease formula. (C) Efferocytosis of pHrodo-labeled dead Jurkat cells by human monocytes from 13 donors in the presence of control mAb (Ctrl) and a humanized anti-CD300A mAb (TNAX103). Left: Representative laser scanning confocal microscopy image. Right: Phagocytic index. Scale bars: 100 µm. (D) Schematic diagram of the generation of the humanized mice by transferring of human CD34⁺ cells to NOG mice with transgenic expression of human GM-CSF and IL-3. These mice were given Ctrl or a humanized anti-CD300A mAb (TNAX103) 2 hours before biIRI. (E) Flow cytometric analysis of CD300A expression on human and mouse CD45⁺ cells in the kidney in the humanized mice after biIRI. (F and G) Plasma NGAL (F) and BUN and Cre (G) before (0) and 48 hours after biIRI in the humanized mice that had been injected i.v. with TNAX103 mAb or Ctrl (n = 5 or 6, respectively). Data are presented as mean ± SEM and pooled from 2 (F and G) and 3 (C) experiments. Student’s t test (C) and 2-way ANOVA (F and G). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.