• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • br caption on next page br Fig CD


    (caption on next page)
    Fig. 1. CD90 is upregulated in pancreatic cancer. (A) Relative mRNA level of CD90 in the pancreas is shown. Data were collected from TCGA (PAAD), GSE15471, GSE28375 and GSE102238 (Renji cohort) (mean ± SEM, unpaired two-tailed Student's t-test). (B) Representative immunohistochemical images of CD90 in a PDAC tissue array. S, stromal cell; T, tumor cell. (C) The expression of CD90 was detected in normal (n = 3), PanINs (n = 13) and PDAC (n = 4) tissues isolated from familial pancreatic cancer in the public GSE43288 database (mean ± SEM, unpaired two-tailed Student's t-test). (D) Flow cytometry analysis of CD90 expression among PDAC cell lines, PDCs and primary tumor Y 27632 (pregated on live+ EpCAM+ cells). (E) GSEA revealed that CD90hi cells (versus control CD90low cells) were positively enriched for Kras signaling, epithelial mesenchymal transition, IL6/JAK/STAT3 signaling and TNFα signaling via NFκB. Data were collected from TCGA (PAAD) database. (F) The association between CD90 transcript level and overall survival in PDAC patients is shown. The analyses were conducted in 179 pancreatic cancer patients from TCGA database (log-rank test P = 0.044).
    mediate the physical interaction between CSCs and macrophages, which could facilitate CSCs to maintain their stem-like state [16]. We speculated that CD90 might serve as an anchor for CD90hi PDAC cells to tether monocytes and derived macrophages in PDAC. First, GSEA from TCGA (PAAD) database revealed that the integrin-associated pathway was enriched in CD90hi cells (Fig. 3A). Next, we analyzed the correla-tions between CD90 and monocyte/macrophage-associated genes, in-cluding CD14, CD68, MRC1 and CD163. Notably, CD90 expression was positively correlated with the expression of CD14, CD68, MRC1 and CD163 in both TCGA database (Fig. 3B) and the Renji cohort (Fig. S3A). The above results suggest that CD90 expression is associated with the existence of monocytes/macrophages.
    In support of this notion, we employed the human monocytic cell line, THP-1. Multiple PDAC cell lines were expanded early in monolayer culture, and then suspended THP-1 cells were added above the cell monolayers for the adhesion assay. CD45, a well-known marker of immune cells, was used to distinguish tumor cells from THP-1 cells (Fig. 3C). Notably, we found that the adhesion capability of PDAC cell lines to THP-1 monocytes was related to their surface CD90 level (Fig. 3D). Moreover, we found that macrophages (CD68) are in close proximity to CD90hi cells in clinical pancreatic tumor tissues by im-munofluorescence staining (Fig. S3B). To further validate the role of CD90 in the adhesion of monocytes/macrophages, we next generated CD90-knockout PANC1 cells using the CRISPR-Cas9 system (Fig. 3E). Knockout of CD90 dramatically reduced the adhesion to THP-1 mono-cytes in PANC1 cells (Fig. 3F) but had little effect on cell proliferation (Fig. S3C). The above results suggest that CD90 could act as an anchor for monocyte/macrophage adhesion in PDAC cells.
    3.4. PDAC cells promote immunosuppressive features of monocytes/ macrophages
    To explore the crosstalk between PDAC cells and attached mono-cytes/macrophages, we generated a coculture system with PDAC and THP-1 cells (Fig. 4A). After coculture for 36 h, we sorted the primed THP-1 cells for quantitative RT-PCR analysis. Surprisingly, we found that the expression of SHH, IL-6, IL8 and IL-10 in THP-1 cells was up-regulated following coculture with PDAC cells (Fig. 4B). Among them, IL-6 and IL-10 are related to the immunosuppressive properties of monocytes/macrophages. Furthermore, PBMCs (peripheral blood mononuclear cells) were treated with conditioned medium from human PDAC cells and then harvested 5 days later for flow cytometry analyses. The ratio of CD14+ monocytes was strongly elevated under PDAC conditioned medium treatment (Fig. 4C). Moreover, primed CD14+ monocytes acquired an immunosuppressive feature characterized by significantly reduced expression of HLA-DR (Fig. 4D) and higher ex-pression of IL-6 and IL-10 (Fig. 4E). To further determine the im-munosuppressive feature of PDAC-primed monocytes, we cultured CD4+ and CD8+ T cells with naïve THP-1 or PDAC-primed THP-1 cells. In this functional assay, we found that the proliferation of T cells (both CD4+ or CD8+ T cells) was significantly impaired when cocultured with PDAC-primed THP-1 cells compared to naïve THP-1 cells (Fig. 4F). Furthermore, by immunostaining of CD90 together with monocyte/ macrophage and T cell markers, we found that high expression of CD90 on PDAC cells was accompanied by abundant macrophages (CD68+) and few CD8+ T cells (Fig. 4G). Thus, PDAC cells trapped monocytes/ 
    macrophages by the CD90/CD11b (MAC-1) axis and reprogrammed them into immunosuppressive states.