Functional and spatial proteomics profiling reveals intra- and intercellular signaling crosstalk in colorectal cancer

Abstract

Precision oncology approaches for patients with colorectal cancer (CRC) continue to lag behind other solid cancers. Functional precision oncology-a strategy that is based on perturbing primary tumor cells from cancer patients-could provide a road forward to personalize treatment. We extend this paradigm to measuring proteome activity landscapes by acquiring quantitative phosphoproteomic data from patient-derived organoids (PDOs). We show that kinase inhibitors induce inhibitor- and patient-specific off-target effects and pathway crosstalk. Reconstruction of the kinase networks revealed that the signaling rewiring is modestly affected by mutations. We show non-genetic heterogeneity of the PDOs and upregulation of stemness and differentiation genes by kinase inhibitors. Using imaging mass-cytometry-based profiling of the primary tumors, we characterize the tumor microenvironment (TME) and determine spatial heterocellular crosstalk and tumor-immune cell interactions. Collectively, we provide a framework for inferring tumor cell intrinsic signaling and external signaling from the TME to inform precision (immuno-) oncology in CRC.

Keywords: Cancer; Cancer systems biology; Proteomics.

Graphical abstract

Figure 1

Schematic outline of the overall concept used in this study Multi-modal profiling and multi-omic profiling of tumor specimens and PDOs in a cohort of CRC patients. See also Table S1.

Figure 2

Proteogenomic analysis of PDOs from CRC patients (A) Genetic profiles of the PDOs ordered according to the mutational load (ML). MMR: mismatch-repair. ML: mutational load. CNV: copy number variation. (B) Analysis of the cancer pathways of the PDOs using bulk RNA-seq data and PROGENy. The pathway activity scores are z-scaled and clustered hierarchically by euclidean distance and complete linkage. (C) Pathway analysis of the hallmark gene sets from MSigDB of the PDOs using proteomic data (SWATH-MS). The heatmap shows z scores of enrichment scores derived from Gene Set Variation Analysis (GSVA) and clustered hierarchically by Pearson correlation as distance metric and complete linkage. (D) Correlation analysis between RNA-seq data and proteomics data. The histogram shows gene-wise Pearson correlation between transcriptome and proteome levels. Denoted are driver genes (black) and immune-related genes (red). The average gene-wise Pearson correlation is 0.29 (dashed line). (E) Protein complexes ranked according to the co-abundance observed for complex members. Shaded areas, left: stable complexes (top 25%), right: variable complexes (bottom 25%). MCM: mini chromosome maintenance. COP9: constitutive photomorphogenesis 9. (F) Variance of the protein levels of the 26S proteasome subunits across all PDOs.

Figure 3

Functional profiling experiments of the PDOs with targeted drugs (A) PCA of the phosphoproteomic data. (B) UpSet plot of regulated phosphopeptides (|log2FC|>1, FDR<0.05) following treatment with specific kinase inhibitors. (C) Heatmap of normalized enrichment scores (NESs) of phosphorylation signatures from PTMSigDB and SIGNOR with at least five phosphorylation sites representing changes in kinase activities following treatment of PDOs with specific kinase inhibitors or TNFα (FDR<0.05), clustered by complete linkage of Euclidean distances. Significant changes and mutations are highlighted with circles and squares, respectively. (D) Normalized enrichment scores (NESs) of phosphorylation signatures representing changes in pathway activities following treatment of PDOs with specific kinase inhibitors or TNFα (FDR<0.05, database and number of phosphorylation sites shown in brackets). Significant changes are highlighted with black circles.

Figure 4

Comparative analysis of the kinase network topologies for the perturbed PDOs (A) Visual representation of the reconstructed kinase networks. Highlighted in color are kinases directly targeted by inhibitors. (B) Eigenvector and degree centrality measures of kinase nodes in the networks shown in (A). Color indicates the number of subgraphs that share a particular node.

Figure 5

Single-cell analysis of PDOs from CRC patients (A) UMAP plot of batch-corrected scRNA-seq dataset from all organoids, colored by cell type. RNA velocity vectors are projected on top of the UMAP plot. (B) UMAP plot from (A) colored by gene expression (log(CPM)) of the markers for stem cells (LGR5), WNT target (AXIN2), goblet cells (TFF3), and enterocytes (FABP1). (C) Cellular composition of the PDOs as measured by scRNA-seq. (D) Analysis of cancer pathways activation in specific epithelial cell types using PROGENy. (E) qPCR measurements represented in a heatmap of stem cell and differentiation gene markers following treatments with different kinase inhibitors for 72 h. Each drug treatment was performed in triplicates. Fold change in expression of target genes was calculated by 2^−ΔΔCT method using DMSO control for normalization, and GAPDH as an endogenous control. ∗p < 0.01; Gene expression fold change values were tested for normality using Shapiro-Wilk test, which showed no deviation from normality. Differences in mean fold change between treated and control were computed by one-way ANOVA with post hoc Dunnet’s test. The resulting p values were corrected for false discovery rate (Benjamini-Hochberg) for the number of target genes.

Figure 6

Spatial proteomics of tumor samples using imaging mass cytometry (A) Cellular composition of the TME using 41 cell phenotypes from six tumor tissues from the respective patients. (B) Cell densities of CD8⁺, CD4⁺, CD45RO⁺, and Tregs. (C) Cell densities of PD1⁺ tumor cells and PD-L1⁺ immune cells. (D) Example subsection (200 × 200 μm) of cell neighborhood analysis using Voronoi diagrams. Upper panel: original image. Lower panel: map following cell phenotype identification and building of Voronoi diagrams. (E) Example subsection (650 × 460μm) of the cell neighborhood analysis for PD1⁺ tumor cells and PD-L1⁺ immune cells interactions in CRC13 (left) and CRC03 (right).