Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: A model to dissect early neoplastic states of pancreatic ductal cells. A Scheme of stem cell-derived pancreatic duct-like organoid (PDLO) sampling for single-cell RNA sequencing (scRNA-seq). B Uniform manifold approximation and projection (UMAP) showing cells of different time points after doxycycline (Dox) treatment. C UMAP of KRT19 expression. D Velocity streamlines representing predicted trajectories within the UMAP. E Ridge plot showing endogenous KRAS (KRAS wt ) and ectopic KRAS G12D expression within different scRNA-seq samples. F UMAPs showing module scores of KRAS-dependent pathways. G Violin plots showing module score of a neoplasia signature (top) and TFF1 expression (bottom) in PDLOs over time of Dox treatment. Unpaired t-test: ****, p < 0.0001; ns, p > 0.05. H Gene-Concept Networks (cnet plots) illustrating KEGG pathway terms and related genes, based on transcription factor (TF) target genes. I Gene ontology (GO) analysis of genes derived from open chromatin regions of assay for transposase-accessible chromatin sequencing (ATAC-seq).

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: Derivative Assay, Sampling, Single Cell, RNA Sequencing, Expressing, Sequencing

Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: KRAS G12D -dependent grouping of cells reveals niche preparatory signatures. A KRAS G12D expression of K-means clusters visualized as median with corresponding extremes. B UMAP illustration of K-means-based KRAS clusters. C Paga velocity graph showing connectivity of KRAS clusters. D UMAP of integrated scRNA-seq data from PDLOs and PDAC [33]. E GO terms related to ECM remodeling and inflammation on DEGs from KRAS high and KRAS intermediate cluster compared to KRAS low . F Heatmap showing log2-fold change values of selectively chosen ECM and inflammation-related genes from DEGs between KRAS high and KRAS intermediate clusters compared to KRAS. low . G Confusion matrix showing predicted and real classification of KRAS clusters on the validation data set (35% of the entire data set) using a deep learning algorithm. H Proportions and I interaction strength of different cell types within PDAC transcriptomes when tumor samples were grouped by KRAS classifiers derived from the deep learning algorithm. Two-way ANOVA: *, p < 0.05; ***, p < 0.001; ****, p < 0.000533

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: Expressing, Biomarker Discovery, Derivative Assay

Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: KRAS G12D expression induces T cell shielding and activates stellate cells towards a CAF phenotype. A Overview images of human PanIN lesions of two donors, stained and imaged for 24 markers via a COMET highplex device, n = 1. Scale bar = 500 μm. B Representative image details of selected stromal (top) and immune cell markers (bottom) of human PanIN lesions subjected to 24-plex immunofluorescence staining. Scale bar = 50 μm. C t-SNE plots illustrating the distribution of indicated markers in human PanIN lesions from 24-plex immunofluorescence staining. D Representative images and respective quantification of T cells infiltrating matrix-embedded organoids after 72 h of co-culture. Mean ± standard deviation (SD) of 3 independent experiments, including a total of 5 different T cell donors. Unpaired t-test ****, p < 0.0001. E Representative images and respective quantification of T cells infiltrating matrix-embedded PDLOs within the cancer-on-chip system. Mean ± SD, n = 1 with 3 chips/condition. Scale bar = 200 μm. F Percentage of T cells migrated towards PDLO conditioned media. Mean ± standard error of mean (SEM) of 4 independent experiments. Unpaired t-test *, p < 0.05. G Proportions of the T cell subpopulations after culturing in PDLO-conditioned media for 72 h were determined by flow cytometry. Mean ± SD, n = 2. H Representative images of PDLOs co-cultured with GFP-labeled pancreatic stellate cells (PaSCs) and respective quantification of the total green area over time of live-cell imaging. Mean ± SEM, n = 3. Unpaired t-test ***, p < 0.001; ****, p < 0.0001. Scale bar = 400 μm. I + J Relative expression of CAF-related marker genes in PaSCs after indirect co-culture with KRAS G12D induced PDLOs (left) or culturing in respective PDLO conditioned media for 72 h (right). The mRNA expression was normalized to GAPDH , and expression is shown relative to the untreated conditions (-Dox). Mean, n = 3, unpaired t-test**, p < 0.01. K Flow cytometry-based analysis of cells positive for αSMA and IL6 after culturing PaSCs in PDLO conditioned media (CM) for 72 h. Mean ± SEM, n = 4 for aSMA, n = 3 for IL6, unpaired t-test **, p < 0.01

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: Expressing, Staining, Immunofluorescence, Co-Culture Assay, Standard Deviation, Flow Cytometry, Cell Culture, Labeling, Live Cell Imaging, Marker

Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: TNFα reduces T cell migration and activates pro-tumorigenic fibroblasts. A Scatter plot showing predicted interaction strengths of different cell types derived from pancreatic tissue of healthy donors with and without PanIN lesions and PDLOs. B Signaling patterns of selected pathways in different cell types derived from pancreatic tissue of healthy donor with and without PanIN lesions and PDLOs. C + D Heatmap showing the interaction probability of ligand-receptor (LR) pairs between PDLOs and cell types derived from pancreatic tissue of healthy donors with and without PanIN lesions (healthy microenvironment, left) and cell types derived from PDAC tissue (tumor microenviroment, right). Only ligand-receptor pairs are shown, which are predicted to originate from either KRAS high or KRAS intermediate but not from KRAS low cells. Overlapping LR-pairs between both analyses are highlighted in bold. E TNF signaling patterns in different cell types derived from pancreatic tissue of healthy donors with and without PanIN lesions and PDLOs. F Circle plot showing predicted interaction strength of the ligand-receptor pair TNF-TNFRSF1A between different cell types derived from pancreatic tissue of healthy donorswith and without PanIN lesions and PDLOs. G Core enriched genes from scRNA-seq data set of GSEA against the HALLMARK_TNFA_SIGNALING_VIA_NFKB gene set and corresponding log2FC values from the ATAC-seq data set. H TNFα concentrations in PDLO-derived conditioned media from 5 independent pancreatic differentiation experiments (left) and relative levels normalized to the untreated -Dox control (right). Mean ± SD. Paired t-test *, p < 0.05. I Percentage of migrated T cells towards PDLO conditioned media ± TNFa/infliximab. Mean ± SEM, n = 4. One-way ANOVA *, p < 0.05; ***, p < 0.001. J t-distributed stochastic neighbor embedding (tSNE) plot showing murine fibroblasts untreated or treated with TNFα for 72 h before subjected to CyTOF analysis (left) and the respective clusters (right). J Proportions of cells per cluster from the t-SNE plot. K Log2FC values of markers from cluster 13 compared to all other clusters

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: Migration, Derivative Assay, Control

Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: Single-cell RNA-seq of PDLO bi-cultures with T cells and pancreatic stellate cells. A Schematic overview of workflow to generate samples for scRNA-seq after co-culture of PDLOs with either T cells or stellate cells. (Scheme was generated using BioRender). B UMAP showing PDLOs, T cells and stellate cells from the scRNA-seq analysis. C Proportions of PDLOs, T cells and stellate cells within the different culture conditions. D Box plot showing the module scores for T cell activation (right), T cell migration (middle) and T cell exhaustion (right) for CD8 + T cell subset, visualized as median with corresponding extremes. Wilcoxon test **, p < 0.01; ns, p > 0.05. E Proportions of CD8 and CD4 cells within the different treatment conditions. F Significance of HALLMARK gene set enrichment showing the top terms (p.adj < 0.05) using DEGs from the indicated comparisons in all T cells. (IFNG Response = Interferon Gamma Response, Unfolded Protein = Unfolded Protein Response, TNF-α via NF-κB = TNF-α Signaling via NF-κB, Cholesterol = Cholesterol Homeostasis, Androgen = Androgen Response, ROS Pathway = Reactive Oxygen Species Pathway, IFNA = Interferon Alpha Response, OXPHOS = Oxidative Phosphorylation). G Heatmap showing log2FC values of DEGs related to mTORC1 Signaling, Cholesterol Homeostasis, ROS and OXPHOS, from indicated comparisons within all T cells (pan), CD8 or CD4 subsets. H Viability of PDLOs when co-cultured with T cells assessed by flow cytometry. I Proportions of myCAF- and iCAF-like stellate cells annotated by module scores using the gene signatures from Elyada et al. . J Dot plot showing module scores of different CAF subpopulations using the top 100 DEGs from Toa et al. . K Box plot showing module scores for the stellate cells using the top 10 DEGs for classical myCAFs and iCAFs from Toa et al. , visualized as median with corresponding extremes. Wilcoxon test ***, p < 0.001; ns, p > 0.05. L Heatmap showing log2FC values of selectively chosen cell cycle and inflammation-related DEGs within the stellate cells when comparing one condition to the other two conditions. M Violin plot showing module score of stromal cells from Peng et al. using the top 100 DEGs of stellate cells co-cultured with PDLOs. N Scatter plot showing predicted interaction strengths of the different cell types within the co-cultures. O TNF signaling patterns in the different cell types within the co-cultures. P Circle plot showing predicted interaction strength of the ligand-receptor pair TNF-TNFRSF1A (left) and TNF-TNFRSF1B (right) between the different cell types within the co-cultures. Q Venn diagram showing overlapping ligand-receptor pairs (left) and respective ligands only (right) when comparing the cell–cell communications analysis between PDLOs with pancreatic tissue of healthy donors , PDAC tissue and the in vitro co-culture. For this comparison, only ligand-receptor pairs were used, which were predicted to originate from either KRAS high or KRAS intermediate but not from KRAS low cells, and receiving cell populations (expressing the corresponding receptor) were limited to T cells, stellate cells, and fibroblasts

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: Single Cell, RNA Sequencing, Co-Culture Assay, Generated, Activation Assay, Migration, Phospho-proteomics, Cell Culture, Flow Cytometry, In Vitro, Comparison, Expressing

Journal: Molecular Cancer

Article Title: An oncogenic KRAS-driven secretome involving TNFα promotes niche preparation prior to pancreatic cancer onset

doi: 10.1186/s12943-025-02541-1

Figure Lengend Snippet: A Representative RNAscope images combined with E-cadherin immunofluorescence showing Tnf-α mRNA puncta in epithelial cells of PanIN lesions from Ptf1a-Cre;Kras LSL−G12D /wt (KC) mice aged 10, 20, and 30 weeks. Subcellular Tnf-α spots indicate discrete transcript signals, while erythrocytes display non-specific pan-staining artifacts. Subcellular Tnf-α expression was, in total, quantified in 4553, 16,740, and 30,084 epithelial cells for the respective 10, 20, and 30 week time points. B Respective quantification of ( A ). The upper bar graph depicts the percentage of epithelial cells harboring ≥ 1 Tnf-α spot, the lower bar graph shows the detailed distribution of Tnf-α spot counts per cell (in %). Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple-comparison test; n = number of PanIN regions analyzed across different mice; C RNAscope–immunofluorescence images of human PanIN-containing pancreatic tissues illustrating TNFA expression within epithelial compartments; D Pathological diagnosis of the 80 patients with IPMN and IPMN-derived invasive PDAC included in the final analysis; E Lesion and tumor distribution of the investigated 80 patients. F Tumor stage distribution of the 28 patients with IPMN-derived invasive PDAC; G Levels of TNF-α in cyst fluid stratified according to the degree of dysplasia; H Comparison of cyst fluid TNF-α levels between patients with low-grade dysplasia and those with high-grade dysplasia or minimally invasive cancer (T1a only); I Comparison of cyst fluid TNF-α levels between patients with low-grade dysplasia and those with high-grade dysplasia or invasive carcinoma arising from IPMN. Statistic test: one tailed Mann Whitney

Article Snippet: For the endogenous KRAS locus, a commercial KRAS FAM target probe assay mix was used (Hs02739788_cn, Thermo Fisher Scientific) which recognizes a region within the 3’UTR of exon 5.

Techniques: RNAscope, Immunofluorescence, Staining, Expressing, Comparison, Biomarker Discovery, Derivative Assay, One-tailed Test, MANN-WHITNEY

Journal: bioRxiv

Article Title: Phosphorylation Protects Oncogenic RAS from LZTR1-Mediated Degradation

doi: 10.64898/2026.01.07.698128

Figure Lengend Snippet: A ) CRISPR screening strategy to identify regulators of KRAS protein stability. B ) Volcano plot of average KRAS stability scores (n=3). Significant hits for genes that either decrease (pink) or increase (purple) KRAS expression are indicated. Callouts represent genes identified as RAS interaction partners by mass spectrometry . C ) Venn diagram of overlapping genes from KRAS stability screen (≤-0.8 KRAS stability score and P≤0.05), RAS BioID2 proteomics (≥1.0 log2 enrichment vs. control, from ref: ), and genes essential in RAS-dependent MM cell lines (≤-1.0 CSS, from ref: ). D ) Western blot analysis of RAS, PPP1R2, and GAPDH 3 days after transduction with control shRNA (shCTRL) or PPP1R2-targeting shRNAs in XG7, RPMI 8226, and MM.1S MM lines, n=3. E ) PPP1R2-BioID2 enrichment over empty vector in RPMI 8226 cells. F ) Western blot analysis of RAS, PPP1R2, PP1C, and GAPDH 3 days after transduction with shCTRL, shPPP1R2.1, and/or ectopic expression of DN PP1C, n=3. G ) Comparison of protein expression levels between KRAS (x-axis) and average PP1C (PPP1CA, PPP1CB, PPP1CC, and PPP1CC;PPP1CB) in 115 MM patient tumors. Display line is simple linear regression; R 2 =0.1593, P<0.0001. I ) Model of PPP1R2 and PP1C regulation of KRAS protein expression. Under a normal state, PPP1R2 inhibits PP1C activity. Following PPP1R2 disruption, PP1C activity reduces KRAS levels. In contrast, PP1C disruption increases KRAS expression.

Article Snippet: Briefly, 150 ng of KRAS gene fragment was combined with 1 μL of SnaBI-linearized BioID-2 vector and 4 μL of 2× NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs) and incubated for 1 h at 50 °C.

Techniques: CRISPR, Expressing, Mass Spectrometry, Control, Western Blot, Transduction, shRNA, Plasmid Preparation, Comparison, Activity Assay, Disruption