Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: A UMAP visualization of the scRNA-seq profiles showing ADSC subpopulations. B Heatmap of marker genes for CXCL14 + ADSC and CD55 + ADSC cells. C Density maps of CD55 and CXCL14 expression in ADSC subpopulations. D , E Gene Ontology enrichment analysis between the CXCL14 + ADSC and CD55 + ADSC subgroups (biological processes shown). F Violin plots comparing enrichment scores for “positive chemotaxis” and “positive regulation of vascular development” gene sets between CXCL14 + ADSCs and CD55 + ADSCs ( n = 10 patients). Box plots show median, 25th–75th percentiles (box), and 5th–95th percentiles (whiskers). The width of violin plot represents the kernel probability density of the data at different values. G Immunofluorescence staining of CD55 + ADSC (CD142 + CD55 + ) and CXCL14 + ADSC (CD142 + GPC3 + ) in PVAT. Scale bar = 25 μm. H Flow cytometry gating strategy and quantification of CD55 + ADSC (CD45 − CD142 + CD55 + ) and CXCL14 + ADSC (CD45 − CD142 + GPC3 + ) in CAS ( n = 14) vs. control ( n = 7) groups. I Immunofluorescence staining of capillaries (Isolectin B4 + ) and CD55 + ADSC (CD142 + CD55 + ) in PVAT. Scale bar = 30 μm. J Correlation of capillary-positive area (Isolectin B4 + ) between ADSC positive area (CD142 + ) and CD55 + ADSC positive area (CD142 + CD55 + ). K , L Correlation of capillary-positive area (CD31 + ) between ADSC ( n = 84) positive area and CD55 + ADSC ( n = 83) positive area. M Quantification of CD55 + ADSC positive area in asymptomatic ( n = 40) v.s. symptomatic ( n = 49) CAS, and in those with ( n = 31) v.s. without ( n = 58) stroke history. N Quantification of CD31-positive area in asymptomatic ( n = 40) v.s. symptomatic ( n = 53) CAS, and in those with ( n = 29) v.s. without ( n = 71) stroke history. Two-sided Wilcoxon rank-sum test with Benjamini–Hochberg correction was used for analyses in ( F ). Two-tailed unpaired t -test was used for analyses in ( H ). Two-tailed Spearman rank-order correlation analysis with linear regression in ( J – L ). Line: linear regression; shaded area: 95% confidence interval. Two-tailed Mann–Whitney test was used for analyses in ( M ) and ( N ). Representative immunofluorescence images of PVAT from 10 ( G ) and 100 ( I ) patients/samples. Error bars show mean ± SD. Data are presented as individual samples in ( H ) and ( J – N ). UMAP uniform manifold approximation and projection, CAS carotid artery stenosis, PVAT perivascular adipose tissue, ADSC adipose-derived stem cell. Source data are provided as a Source Data file.

Article Snippet: Experimental conditions included a serum-free negative control, a normal medium control, conditioned medium from CD55 + ADSCs (CD55 + ADSC-CM) and CXCL14 + ADSCs (CXCL14 + ADSC-CM), as well as CD55 + ADSC-CM containing 25 nM of the FGFR1 inhibitor PD173074 (HY-10321, MCE, China).

Techniques: Marker, Expressing, Chemotaxis Assay, Immunofluorescence, Staining, Flow Cytometry, Control, Two Tailed Test, MANN-WHITNEY, Derivative Assay

Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: A Schematic of carotid partial ligation and intervention experimental procedures in C57 mice and Balb/c nude mice. This schematic was created in BioRender. Chen, J. (2026) https://BioRender.com/4l6sgra . B , C Evaluation of carotid artery remodeling following carotid ligation and murine ADSC transplantation in C57 mice. B Representative EVG staining. C Quantification of neointima and media remodeling across 8 sections (1400 μm span) for CXCL14 + ADSC ( n = 5), CD55 + ADSC ( n = 6), and control ( n = 4) groups. D , E Evaluation of capillary formation (CD31) and ADSC engraftment (GFP) after carotid ligation and murine ADSC transplantation in C57 mice. D Representative images. Scale bar = 50 μm. E Quantification of CD31 + area in ligated arteries. Groups and n values as in ( C ). F – H Vascular remodeling in nude mice after ligation and human ADSC transplantation. F Representative images of Oil Red O staining. Scale bar = 100 μm. G Representative images of EVG staining. Groups and n values as in ( H ). H Quantification of neointima remodeling based on EVG staining (across 8 sections spanning 1400 μm) for CXCL14 + ADSC ( n = 6), CD55 + ADSC ( n = 5), and control ( n = 5) groups. I – L Immunofluorescence assessment of angiogenesis and human ADSC engraftment in nude mice. I , J Representative images of capillary formation (Isolectin B4), human endothelial cell formation (human CD31), and ADSC engraftment (GFP). Scale bar = 100 μm. K Quantification of Isolectin B4 positive area. L Quantification of human CD31-positive area. Groups and n values as in ( H ). M t-SNE visualization and evolutionary tree of lineages from CD55 + ADSC and CXCL14 + ADSC of PVAT to CD31 + endothelial cells of plaque from the same patient with carotid stenosis. Two-tailed two-way ANOVA followed by Tukey’s multiple comparisons test was used for analyses in ( C ) and ( H ). Two-tailed one-way ANOVA followed by Tukey’s multiple comparisons test was used for analyses in ( E , K , and L ). Data are presented as individual samples in ( C , E , H , and K – L ). Data are presented as the mean ± SD. CAS carotid artery stenosis, PVAT perivascular adipose tissue, ADSC adipose-derived stem cell, ns no significance. Source data are provided as a Source Data file.

Article Snippet: Experimental conditions included a serum-free negative control, a normal medium control, conditioned medium from CD55 + ADSCs (CD55 + ADSC-CM) and CXCL14 + ADSCs (CXCL14 + ADSC-CM), as well as CD55 + ADSC-CM containing 25 nM of the FGFR1 inhibitor PD173074 (HY-10321, MCE, China).

Techniques: Ligation, Transplantation Assay, Staining, Control, Immunofluorescence, Two Tailed Test, Derivative Assay

Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: During carotid stenosis, the expanded ADSC population includes distinct functional subsets. The CD55 + ADSC subset promotes plaque instability by migrating into lesions, potentially differentiating into endothelial cells, and secreting FGF2 to concurrently stimulate angiogenesis and a synthetic phenotypic switch in smooth muscle cells, resulting in neointimal hyperplasia. Conversely, the CXCL14 + ADSC subset functions in inflammatory recruitment by leveraging pathways like CXCL12-CXCR4 to attract macrophages and other immune cells. Furthermore, extensive cell-cell interactions within PVAT, particularly between these ADSCs and lymphocytes/myeloid cells, orchestrate a synergistic effect that amplifies chemotaxis and angiogenesis, ultimately accelerating the formation of unstable, clinically significant plaques that predispose to stroke. This schematic was created in BioRender. Chen, J. (2026) https://BioRender.com/uf4cxuu .

Article Snippet: Experimental conditions included a serum-free negative control, a normal medium control, conditioned medium from CD55 + ADSCs (CD55 + ADSC-CM) and CXCL14 + ADSCs (CXCL14 + ADSC-CM), as well as CD55 + ADSC-CM containing 25 nM of the FGFR1 inhibitor PD173074 (HY-10321, MCE, China).

Techniques: Functional Assay, Chemotaxis Assay

Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: A UMAP visualization of the scRNA-seq profiles showing ADSC subpopulations. B Heatmap of marker genes for CXCL14 + ADSC and CD55 + ADSC cells. C Density maps of CD55 and CXCL14 expression in ADSC subpopulations. D , E Gene Ontology enrichment analysis between the CXCL14 + ADSC and CD55 + ADSC subgroups (biological processes shown). F Violin plots comparing enrichment scores for “positive chemotaxis” and “positive regulation of vascular development” gene sets between CXCL14 + ADSCs and CD55 + ADSCs ( n = 10 patients). Box plots show median, 25th–75th percentiles (box), and 5th–95th percentiles (whiskers). The width of violin plot represents the kernel probability density of the data at different values. G Immunofluorescence staining of CD55 + ADSC (CD142 + CD55 + ) and CXCL14 + ADSC (CD142 + GPC3 + ) in PVAT. Scale bar = 25 μm. H Flow cytometry gating strategy and quantification of CD55 + ADSC (CD45 − CD142 + CD55 + ) and CXCL14 + ADSC (CD45 − CD142 + GPC3 + ) in CAS ( n = 14) vs. control ( n = 7) groups. I Immunofluorescence staining of capillaries (Isolectin B4 + ) and CD55 + ADSC (CD142 + CD55 + ) in PVAT. Scale bar = 30 μm. J Correlation of capillary-positive area (Isolectin B4 + ) between ADSC positive area (CD142 + ) and CD55 + ADSC positive area (CD142 + CD55 + ). K , L Correlation of capillary-positive area (CD31 + ) between ADSC ( n = 84) positive area and CD55 + ADSC ( n = 83) positive area. M Quantification of CD55 + ADSC positive area in asymptomatic ( n = 40) v.s. symptomatic ( n = 49) CAS, and in those with ( n = 31) v.s. without ( n = 58) stroke history. N Quantification of CD31-positive area in asymptomatic ( n = 40) v.s. symptomatic ( n = 53) CAS, and in those with ( n = 29) v.s. without ( n = 71) stroke history. Two-sided Wilcoxon rank-sum test with Benjamini–Hochberg correction was used for analyses in ( F ). Two-tailed unpaired t -test was used for analyses in ( H ). Two-tailed Spearman rank-order correlation analysis with linear regression in ( J – L ). Line: linear regression; shaded area: 95% confidence interval. Two-tailed Mann–Whitney test was used for analyses in ( M ) and ( N ). Representative immunofluorescence images of PVAT from 10 ( G ) and 100 ( I ) patients/samples. Error bars show mean ± SD. Data are presented as individual samples in ( H ) and ( J – N ). UMAP uniform manifold approximation and projection, CAS carotid artery stenosis, PVAT perivascular adipose tissue, ADSC adipose-derived stem cell. Source data are provided as a Source Data file.

Article Snippet: The primary antibodies used included those against PDGFRA (1:200, ab203491, EPR22059-270, Abcam, USA), PDGFRB (1:200, A19531, ARC0009, Abclonal, China), CD142 (1:200, ab228968, EPR22548-232, Abcam, USA), GPC3 (1:200, A11686, ARC0675, Abclonal, China), CD55 (1:200, A22324, ARC52244 , ABclonal, China) and CD31 antibody (1:200, GB153151 , SB166, Servicebio, China).

Techniques: Marker, Expressing, Chemotaxis Assay, Immunofluorescence, Staining, Flow Cytometry, Control, Two Tailed Test, MANN-WHITNEY, Derivative Assay

Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: A Schematic of carotid partial ligation and intervention experimental procedures in C57 mice and Balb/c nude mice. This schematic was created in BioRender. Chen, J. (2026) https://BioRender.com/4l6sgra . B , C Evaluation of carotid artery remodeling following carotid ligation and murine ADSC transplantation in C57 mice. B Representative EVG staining. C Quantification of neointima and media remodeling across 8 sections (1400 μm span) for CXCL14 + ADSC ( n = 5), CD55 + ADSC ( n = 6), and control ( n = 4) groups. D , E Evaluation of capillary formation (CD31) and ADSC engraftment (GFP) after carotid ligation and murine ADSC transplantation in C57 mice. D Representative images. Scale bar = 50 μm. E Quantification of CD31 + area in ligated arteries. Groups and n values as in ( C ). F – H Vascular remodeling in nude mice after ligation and human ADSC transplantation. F Representative images of Oil Red O staining. Scale bar = 100 μm. G Representative images of EVG staining. Groups and n values as in ( H ). H Quantification of neointima remodeling based on EVG staining (across 8 sections spanning 1400 μm) for CXCL14 + ADSC ( n = 6), CD55 + ADSC ( n = 5), and control ( n = 5) groups. I – L Immunofluorescence assessment of angiogenesis and human ADSC engraftment in nude mice. I , J Representative images of capillary formation (Isolectin B4), human endothelial cell formation (human CD31), and ADSC engraftment (GFP). Scale bar = 100 μm. K Quantification of Isolectin B4 positive area. L Quantification of human CD31-positive area. Groups and n values as in ( H ). M t-SNE visualization and evolutionary tree of lineages from CD55 + ADSC and CXCL14 + ADSC of PVAT to CD31 + endothelial cells of plaque from the same patient with carotid stenosis. Two-tailed two-way ANOVA followed by Tukey’s multiple comparisons test was used for analyses in ( C ) and ( H ). Two-tailed one-way ANOVA followed by Tukey’s multiple comparisons test was used for analyses in ( E , K , and L ). Data are presented as individual samples in ( C , E , H , and K – L ). Data are presented as the mean ± SD. CAS carotid artery stenosis, PVAT perivascular adipose tissue, ADSC adipose-derived stem cell, ns no significance. Source data are provided as a Source Data file.

Article Snippet: The primary antibodies used included those against PDGFRA (1:200, ab203491, EPR22059-270, Abcam, USA), PDGFRB (1:200, A19531, ARC0009, Abclonal, China), CD142 (1:200, ab228968, EPR22548-232, Abcam, USA), GPC3 (1:200, A11686, ARC0675, Abclonal, China), CD55 (1:200, A22324, ARC52244 , ABclonal, China) and CD31 antibody (1:200, GB153151 , SB166, Servicebio, China).

Techniques: Ligation, Transplantation Assay, Staining, Control, Immunofluorescence, Two Tailed Test, Derivative Assay

Journal: Nature Communications

Article Title: Perivascular adipose single-cell atlas identifies CD55 + adipose-derived stem cells as vascular remodeling regulators in atherosclerosis

doi: 10.1038/s41467-026-72962-z

Figure Lengend Snippet: During carotid stenosis, the expanded ADSC population includes distinct functional subsets. The CD55 + ADSC subset promotes plaque instability by migrating into lesions, potentially differentiating into endothelial cells, and secreting FGF2 to concurrently stimulate angiogenesis and a synthetic phenotypic switch in smooth muscle cells, resulting in neointimal hyperplasia. Conversely, the CXCL14 + ADSC subset functions in inflammatory recruitment by leveraging pathways like CXCL12-CXCR4 to attract macrophages and other immune cells. Furthermore, extensive cell-cell interactions within PVAT, particularly between these ADSCs and lymphocytes/myeloid cells, orchestrate a synergistic effect that amplifies chemotaxis and angiogenesis, ultimately accelerating the formation of unstable, clinically significant plaques that predispose to stroke. This schematic was created in BioRender. Chen, J. (2026) https://BioRender.com/uf4cxuu .

Article Snippet: The primary antibodies used included those against PDGFRA (1:200, ab203491, EPR22059-270, Abcam, USA), PDGFRB (1:200, A19531, ARC0009, Abclonal, China), CD142 (1:200, ab228968, EPR22548-232, Abcam, USA), GPC3 (1:200, A11686, ARC0675, Abclonal, China), CD55 (1:200, A22324, ARC52244 , ABclonal, China) and CD31 antibody (1:200, GB153151 , SB166, Servicebio, China).

Techniques: Functional Assay, Chemotaxis Assay

Journal: eLife

Article Title: Single-nucleus transcriptional and chromatin accessibility analyses of maturing mouse Achilles tendon uncover the molecular landscape of tendon stem/progenitor cells

doi: 10.7554/eLife.104768

Figure Lengend Snippet: ( A ) Integrated uniform manifold approximation and projection (UMAP) scRNA-seq clustering of cells harvested from 2-week-old and 6-week-old mouse Achilles tendons. ( B ) Dot plot of average gene expression levels of the indicated genes in each scRNA-seq cluster. The size of the dot reflects the percentage of cells in the cluster that express each gene. TC, tenocyte; SP1, tendon stem/progenitor cell_1; CA, cartilage; RB, ribosomal RNA; SP2, tendon stem/progenitor cell_2; LC1, lymphocyte_1; EC, endothelial cell; RBC, red blood cell; SM, smooth muscle cell; PC, proliferating cell; LC2, lymphocytes_2; MTJ, myotendinous junction cell; SW, Schwann cell; VEC, vascular endothelial cell; MC, macrophage. ( C ) Proportions of cells in clusters identified from scRNA-seq. Clusters are colored according to cluster type. ( D ) Volcano plot of gene expression in the SP2 cluster and the identification of candidate TSPC marker genes (red under line). ( E ) Feature plot of Cd55 and Cd248 expression. ( F ) Correlation of gene expression of TSPC candidate genes in 2-week data. ( G ) Violin plots presenting the gene expression changes for a selection of differentially expressed genes.

Article Snippet: Sections of 10 μm in thickness were deparaffinized and activated with citric acid buffer (10 mM sodium citrate, 1 mM EDTA; pH 6.0) at 80°C for 60 min in a decloaking chamber NxGen (BIOCARE Medical, CA, USA) or with 0.1% trypsin/PBS at 37°C for 30 min. After they were blocked with Blocking One solution (Nacalai Tesque, Kyoto, Japan) for 60 min, they were incubated with rabbit anti-CD55 antibody (1:100; A13918, RRID: AB_2760771 , ABclonal, Woburn, MA, USA) or rat anti-CD248 (1:100; MAB7535, RRID: AB_3658829 , R&D Systems, Minneapolis, MN, USA) overnight at 4°C.

Techniques: Gene Expression, Marker, Expressing, Selection

Journal: eLife

Article Title: Single-nucleus transcriptional and chromatin accessibility analyses of maturing mouse Achilles tendon uncover the molecular landscape of tendon stem/progenitor cells

doi: 10.7554/eLife.104768

Figure Lengend Snippet: ( A ) Uniform manifold approximation and projection (UMAP) representation of snRNA-seq differentiation trajectory of tenocytes and TSPC lineage and pseudotime-dependent gene expression changes of Tppp3 , Cd55 , Cd248 , Mkx, and Scx, as inferred using Monocle3. ( B ) UMAP representation of snATAC-seq differentiation trajectory of tenocytes and the TSPC lineage and pseudotime-dependent gene expression changes, as inferred using Cicero. ( C ) Coverage plots of Mkx , Scx , Cd55, and Cd248 . Selected peaks that differ across each cluster are highlighted.

Article Snippet: Sections of 10 μm in thickness were deparaffinized and activated with citric acid buffer (10 mM sodium citrate, 1 mM EDTA; pH 6.0) at 80°C for 60 min in a decloaking chamber NxGen (BIOCARE Medical, CA, USA) or with 0.1% trypsin/PBS at 37°C for 30 min. After they were blocked with Blocking One solution (Nacalai Tesque, Kyoto, Japan) for 60 min, they were incubated with rabbit anti-CD55 antibody (1:100; A13918, RRID: AB_2760771 , ABclonal, Woburn, MA, USA) or rat anti-CD248 (1:100; MAB7535, RRID: AB_3658829 , R&D Systems, Minneapolis, MN, USA) overnight at 4°C.

Techniques: Gene Expression

Journal: eLife

Article Title: Single-nucleus transcriptional and chromatin accessibility analyses of maturing mouse Achilles tendon uncover the molecular landscape of tendon stem/progenitor cells

doi: 10.7554/eLife.104768

Figure Lengend Snippet: ( A ) Immunohistochemical image of 10-week mouse Achilles tendons. Scale bars show 100 µm. CD55 and CD248, green; Hoechst 33342, blue. ( B ) Schema of the in vitro assessment of the capacity of CD55+/CD248+ TSPCs as the differentiation toward tenocytes. ( C ) Colony-forming efficiency of CD55+/CD248+ and CD55-/CD248- (negative) TSPCs. Colonies were stained with crystal violet (n=6). CD55+/CD248+ TSPC exhibited higher clonogenic capacity. Data are presented as means ± scanning electron microscopy (SEM). **p<0.01. ( D ) Morphological changes of CD55+/CD248+ and negative TSPCs after tenogenic induction. ( E ) Quantitative PCR of tendon-related gene expression in CD55+/CD248+ and negative TSPCs after tenogenic induction (n=3). Data are presented as means ± SD. **p<0.01, *p<0.05. ( F ) SEM and transmission electron microscopy (TEM) imaging of artificial tendons derived from CD55+/CD248+ and negative TSPCs. Data are presented as means ± SD. **p<0.01, *p<0.05. ( G ) Proportions of fiber alignment for each artificial tendon (n=4). Data are presented as means ± SEM. ***p<0.005, **p<0.01, *p<0.05. ( H ) Diameter of collagen fiber in each artificial tendon based on TEM imaging (n=4). Data are presented as means ± SD. ***p<0.005. ( I ) Tensile strength (MPa) of each artificial tendon (n=5). Data are presented as means ± SEM. ***p<0.005.

Article Snippet: Sections of 10 μm in thickness were deparaffinized and activated with citric acid buffer (10 mM sodium citrate, 1 mM EDTA; pH 6.0) at 80°C for 60 min in a decloaking chamber NxGen (BIOCARE Medical, CA, USA) or with 0.1% trypsin/PBS at 37°C for 30 min. After they were blocked with Blocking One solution (Nacalai Tesque, Kyoto, Japan) for 60 min, they were incubated with rabbit anti-CD55 antibody (1:100; A13918, RRID: AB_2760771 , ABclonal, Woburn, MA, USA) or rat anti-CD248 (1:100; MAB7535, RRID: AB_3658829 , R&D Systems, Minneapolis, MN, USA) overnight at 4°C.

Techniques: Immunohistochemical staining, In Vitro, Staining, Electron Microscopy, Real-time Polymerase Chain Reaction, Gene Expression, Transmission Assay, Imaging, Derivative Assay

Journal: eLife

Article Title: Single-nucleus transcriptional and chromatin accessibility analyses of maturing mouse Achilles tendon uncover the molecular landscape of tendon stem/progenitor cells

doi: 10.7554/eLife.104768

Figure Lengend Snippet: Quantitative PCR of gene expression in CD55+/CD248+ and negative TSPCs (n=4). Data are presented as means ± scanning electron microscopy (SEM). *p<0.05, ***p<0.005.

Article Snippet: Sections of 10 μm in thickness were deparaffinized and activated with citric acid buffer (10 mM sodium citrate, 1 mM EDTA; pH 6.0) at 80°C for 60 min in a decloaking chamber NxGen (BIOCARE Medical, CA, USA) or with 0.1% trypsin/PBS at 37°C for 30 min. After they were blocked with Blocking One solution (Nacalai Tesque, Kyoto, Japan) for 60 min, they were incubated with rabbit anti-CD55 antibody (1:100; A13918, RRID: AB_2760771 , ABclonal, Woburn, MA, USA) or rat anti-CD248 (1:100; MAB7535, RRID: AB_3658829 , R&D Systems, Minneapolis, MN, USA) overnight at 4°C.

Techniques: Real-time Polymerase Chain Reaction, Gene Expression, Electron Microscopy

Journal: eLife

Article Title: Single-nucleus transcriptional and chromatin accessibility analyses of maturing mouse Achilles tendon uncover the molecular landscape of tendon stem/progenitor cells

doi: 10.7554/eLife.104768

Figure Lengend Snippet: Quantitative PCR of tendon-related genes in CD55+/CD248+ and negative TSPCs after chondrogenic and osteogenic induction (n=4). Data are presented as means ± scanning electron microscopy (SEM). **p<0.01, *p<0.05.

Article Snippet: Sections of 10 μm in thickness were deparaffinized and activated with citric acid buffer (10 mM sodium citrate, 1 mM EDTA; pH 6.0) at 80°C for 60 min in a decloaking chamber NxGen (BIOCARE Medical, CA, USA) or with 0.1% trypsin/PBS at 37°C for 30 min. After they were blocked with Blocking One solution (Nacalai Tesque, Kyoto, Japan) for 60 min, they were incubated with rabbit anti-CD55 antibody (1:100; A13918, RRID: AB_2760771 , ABclonal, Woburn, MA, USA) or rat anti-CD248 (1:100; MAB7535, RRID: AB_3658829 , R&D Systems, Minneapolis, MN, USA) overnight at 4°C.

Techniques: Real-time Polymerase Chain Reaction, Electron Microscopy