Journal: Cancer Gene Therapy

Article Title: NrCAM secreted by endometrial stromal cells enhances the progestin sensitivity of endometrial cancer cells through epigenetic modulation of PRB

doi: 10.1038/s41417-022-00467-0

Figure Lengend Snippet: A Cluster analysis of MPA-regulated genes encoding secretory proteins in ESCs. ESCs were treated with 10 μM MPA or ethanol (EtOH) for 6 h before RNA sequencing. B mRNA levels of NrCAM , BMP2 , WISP1 , and ITGA10 were significantly upregulated in ESCs after MPA treatment. Silencing PR expression with si PGR in ESCs weakened MPA-induced upregulation of these four proteins. ESCs or ESCs-si PGR were treated with or without 10 μM MPA for 6 h. Eleven candidate gene mRNA levels were reevaluated by real-time PCR. C Exogenous NrCAM inhibited EC cell proliferation in a dose-dependent manner. Ishikawa and ECC-1 cells were treated with 0, 1, 10, 100, and 1000 ng/mL NrCAM for 48 h before CCK-8 assays. D Exogenous NrCAM inhibited EC cell proliferation in a time-dependent manner. Ishikawa and ECC-1 cells were treated with 1000 ng/mL NrCAM for 24, 48, and 72 h before CCK-8 assays. E MPA promoted NrCAM protein expression in ESCs in a dose-dependent manner. NrCAM expression was detected by western blotting. ESCs were treated with 0, 5, 10, and 20 μM MPA for 48 h. F MPA promoted NrCAM secretion in ESCs by ELISA. ESCs were treated with MPA at the indicated dose for 48 h (left) or 10 μM MPA for 24, 48, or 72 h (right). The CM extracted from ESCs was collected to measure NrCAM concentration by ELISA. G MPA-induced NrCAM protein expression was attenuated by silencing PGR in ESCs. ESCs or ESCs-si PGR were treated with 10 μM MPA for 48 h before western blotting analysis. H NrCAM and MPA cotreatment had a stronger inhibitory effect on EC cell proliferation than MPA or NrCAM alone. Ishikawa and ECC-1 cells were treated with 1000 ng/mL NrCAM and/or 10 μM MPA for 48 h before CCK-8 assays. I Transfection efficiency of siRNAs targeting NrCAM was confirmed by real-time PCR and western blotting. J The inhibitory effect of ESCs on EC cell proliferation was blocked by silencing NrCAM expression in ESCs. After transfection with si NrCAM or siCtrl for 8 h, ESCs were treated with or without 10 μM MPA for 48 h, then for CM collection. Ishikawa and ECC-1 cells were treated with 10 μM MPA, CM (ESCs-si NrCAM -2) and CM (ESCs-si NrCAM -2 + MPA) for 48 h before CCK-8 assays. * P < 0.05; ** P < 0.01; *** P < 0.001; n.s. not significant.

Article Snippet: Drugs used in this study included MPA (Sigma-Aldrich, St. Louis, MO, USA), recombinant human NrCAM protein (CC04, NovoProtein, Shanghai, China), recombinant human BMP2 protein (C012, NovoProtein), recombinant human WISP1 protein (10442-H08H, Sino Biological, Beijing, China), and recombinant human ITGA10 protein (5895-AB-050, R&D Systems, Minneapolis, MN, USA) at the indicated doses for the indicated periods.

Techniques: RNA Sequencing Assay, Expressing, Real-time Polymerase Chain Reaction, CCK-8 Assay, Western Blot, Enzyme-linked Immunosorbent Assay, Concentration Assay, Transfection

Journal: Nature Communications

Article Title: Data-driven learning how oncogenic gene expression locally alters heterocellular networks

doi: 10.1038/s41467-022-29636-3

Figure Lengend Snippet: Digital cytometry deconvolutes a bulk transcriptomic profile using gene signatures that correspond to different stromal, malignant, and immune cell types. The results estimate the prevalence of the different cell types within the tissue sample, that is the digital cytometry features. By using bulk transcriptomic profiles of defined patient populations, underlying variation in the inferred cellular composition coupled with features associated with a patient sample, such as over-expression of a secreted gene product by malignant cells, can be used to estimate how the heterocellular network is impacted by a genetic alteration intrinsic to the malignant cell using Bayesian Network inference. To illustrate the approach, we focused on malignant cell expression of Cell Communication Network factor 4 (CCN4), a secreted matricellular protein. The resulting directed acyclic graph represents the collective conditional independence among the modeled features, or nodes, of the network.

Article Snippet: The concentration of CCN4 protein in the cell culture media from those wells was assayed using the Human WISP-1/CCN4 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN) to confirm CCN4 knockout.

Techniques: Cytometry, Over Expression, Expressing

Journal: Nature Communications

Article Title: Data-driven learning how oncogenic gene expression locally alters heterocellular networks

doi: 10.1038/s41467-022-29636-3

Figure Lengend Snippet: The nodes of the graph represent features, such as CCN4 gene expression (rectangle), sample attribute (hexagon), or the prevalence of a particular cell type/state (oval). The edges represent inferred causal relationships among the nodes. The black lines with arrow heads represent a positive causal relation while red lines with horizontal bars represent a negative or inhibitory causal relation, where the extent of influence of the parental node is annotated by the number beside the edge. The number included within the node symbol represents the average normalized value of the digital cytometry feature within the dataset with values of all of the parental nodes set to zero. The width of the edge is proportional to the posterior probability of inclusion into the DAG.

Article Snippet: The concentration of CCN4 protein in the cell culture media from those wells was assayed using the Human WISP-1/CCN4 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN) to confirm CCN4 knockout.

Techniques: Expressing, Cytometry

Journal: Nature Communications

Article Title: Data-driven learning how oncogenic gene expression locally alters heterocellular networks

doi: 10.1038/s41467-022-29636-3

Figure Lengend Snippet: A The percentage of live CD45+ cells isolated from tumors generated by inoculating s.c. with WT (red) and CCN4 KO (blue) variants of B16F0 (o and x’s) and YUMM1.7 (□ and +’s) cells, where the log-linear trends are highlighted by dotted lines. CD45+ values were obtained from three different antibody panels that quantified T cells, B/NK cells, and myeloid cells in TIL isolates from each mouse. B A comparison of the ratio of NK cells (black), CD8+ T cells (red), CD4+ T cells (blue), and B cells (green) to live CD45+ TILs in s.c. tumors generated using WT B16F0 and YUMM1.7 cells (mean ± s.d.). C The difference in the mean prevalence of the infiltrating immune cell types was compared when CCN4 is present (WT) versus absent (CCN4 KO) as predicted by digital cytometry from the BRCA (dark gray) and SKCM (light gray) datasets and as observed experimentally using the B16F0 (red) and YUMM1.7 (black) mouse models. D TIL comparison upon CCN4 KO in B16F0 and YUMM1.7 mouse models stratified by NK cells, CD8+ T cells, CD4+ T cells, and B cells (top to bottom) ( n = 7 biologically independent animals for YUMM1.7 and n = 4 biologically independent animals for B16F0 variants and mean ± s.d.). p -values calculated between WT and CCN4 KO pairs using two-sided Student’s t test.

Article Snippet: The concentration of CCN4 protein in the cell culture media from those wells was assayed using the Human WISP-1/CCN4 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN) to confirm CCN4 knockout.

Techniques: Isolation, Generated, Comparison, Cytometry

Journal: Nature Communications

Article Title: Data-driven learning how oncogenic gene expression locally alters heterocellular networks

doi: 10.1038/s41467-022-29636-3

Figure Lengend Snippet: A A comparison of the ratio of CD11c- (black) and CD11c+ (gray) macrophages, Dendritic cells (yellow), CD11c+ MDSC (green), MDSC (blue), and Neutrophils (red) to live CD45+ TILs in s.c. tumors generated using WT B16F0 and YUMM1.7 cells (mean ± s.d.). B The difference in prevalence of the myeloid cell types was compared when CCN4 is present (WT) versus absent (CCN4 KO) as predicted by digital cytometry of the BRCA (dark gray) and SKCM (light gray) data sets and as observed experimentally using the B16F0 (red) and YUMM1.7 (black) mouse models. Macrophages are the only myeloid cell subset inferred from the BRCA and SKCM datasets and are assumed to be related to CD11c+ macrophages in mouse models. C A representative scatter plot of GR1 versus CD11c expression in gated live CD45+ CD11b+ TILs obtained from WT (top) and CCN4 KO (bottom) YUMM1.7 tumors. D – F TIL comparison upon CCN4 KO in B16F0 and YUMM1.7 mouse models stratified by myeloid-derived suppressor cell subsets ( D : MDSC (top) and CD11c+ MDSC (bottom)) and other myeloid cell subsets ( E : CD11c- (top) and CD11c+ (bottom) macrophages, F : neutrophils (top) and dendritic cells (bottom)) ( n = 7 biologically independent animals for YUMM1.7 and n = 4 biologically independent animals for B16F0 variants and mean ± s.d.). p -values calculated between WT and CCN4 KO variants using two-sided Student’s t test.

Article Snippet: The concentration of CCN4 protein in the cell culture media from those wells was assayed using the Human WISP-1/CCN4 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN) to confirm CCN4 knockout.

Techniques: Comparison, Generated, Cytometry, Expressing, Derivative Assay

Journal: Nature Communications

Article Title: Data-driven learning how oncogenic gene expression locally alters heterocellular networks

doi: 10.1038/s41467-022-29636-3

Figure Lengend Snippet: A Expression of genes for transcription factors (left panel - Snai1: red triangle, Snai2: blue diamond, Zeb1: black circle, and Zeb2: gray square) and adhesion proteins (right panel - Cdh1: blue triangle, Cdh2: black circle, Fn1: red square) associated with the epithelial-mesenchymal transition were assayed as a function of time following addition of rmCCN4 to CCN4 KO YUMM1.7 (top row) and CCN4 KO B16F0 cells (bottom row). Colored asterisks indicate whether gene at a particular time point was significantly different than untreated cells, where n = 3 biological independent samples. B The distribution in cell trace staining among live CD4 + (left panel) and CD8 + (right panel) T cells stimulated with α CD3/ α CD28 (AP beads) alone or in the presence of media conditioned by WT B16F0 cells (AP beads + WT TCM), media conditioned by CCN4 KO B16F0 cells (AP beads + CCN4 KO TCM), or with 10 ng/ml of recombinant mouse CCN4 (AP beads + rCCN4). The distribution in the corresponding unstimulated cells (gray) are shown at the bottom. The colored vertical lines indicate the predicted dilution of cell trace staining in each generation based on the unstimulated controls. C Bivariate projection of the weights of genes within the resting (y-axis) and activated (x-axis) NK cell signatures. D Using spleens from C57BL/6 mice that were challenged with YUMM1.7 cells, isolated CD8+ T cells were assayed by in vitro ELISpot for IFN γ expression using variants of the YUMM1.7 cell line as targets (CCN4 KO YUMM1.7 with a blank inducible expression vector and CCN4 KO YUMM1.7 with a CCN4 inducible expression vector). To induce CCN4 expression, these YUMM1.7 variants were also cultured in the absence (−) or presence of doxycycline (+) and quantified following 24 h co-culture. Statistical significance between WT and CCN4 KO variants was assessed using two-way ANOVA followed by Tukey’s multiple comparison ad hoc post-test, where n = 6 biologically independent samples. Results summarized as mean ± s.d.

Article Snippet: The concentration of CCN4 protein in the cell culture media from those wells was assayed using the Human WISP-1/CCN4 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN) to confirm CCN4 knockout.

Techniques: Expressing, Staining, Recombinant, Isolation, In Vitro, Enzyme-linked Immunospot, Plasmid Preparation, Cell Culture, Co-Culture Assay, Comparison