Revealing the transcriptional heterogeneity of organ‐specific metastasis in human gastric cancer using single‐cell RNA Sequencing

Haiping Jiang, Dingyi Yu, Penghui Yang, Rongfang Guo, Mei Kong, Yuan Gao, Xiongfei Yu, Xiaoyan Lu, Xiaohui FanClinical 2022 · DOI

TL;DR

Gastric cancer (GC) is a deadly disease, and understanding how it spreads to different organs is crucial for effective treatment. This study uses single-cell RNA sequencing to map the transcriptional landscape of GC primary tumors and their metastases, revealing distinct cellular subtypes and their roles in metastasis and patient survival. The findings offer new insights into GC's complex biology and suggest potential therapeutic targets.

What Problem Does This Paper Address?

Gastric cancer (GC) is a major global health challenge, with a high mortality rate, largely due to its tendency to metastasize to various organs. While we know that GC can spread to the lymph nodes, liver, peritoneum, and ovaries, the specific molecular mechanisms and cellular players driving this organ-specific metastasis are not well understood. This lack of understanding hinders the development of targeted therapies. Previous studies using bulk RNA sequencing have provided some insights, but they obscure the intricate cellular diversity and heterogeneity within tumors. This paper aims to address this gap by using single-cell RNA sequencing (scRNA-seq) to provide a high-resolution view of the transcriptional landscape of GC primary tumors and their metastases, uncovering the cellular heterogeneity and interactions that drive organ-specific spread.

Methods

The researchers employed single-cell RNA sequencing (scRNA-seq) on 10 fresh human tissue samples from six GC patients, including primary tumors and metastases from different organs (liver, peritoneum, ovary, lymph node).

Key techniques and analyses included:

  • scRNA-seq: To capture the transcriptome of individual cells.
  • Dimensionality Reduction and Clustering: Using t-SNE and UMAP to visualize and group cells based on their gene expression profiles.
  • Cell Type Annotation: Identifying major cell lineages (epithelial, stromal, immune cells like T cells, B cells, NK cells, and myeloid cells) using canonical marker genes.
  • Subcluster Analysis: Further dissecting the heterogeneity within each major cell type to identify distinct cellular subtypes.
  • Differential Gene Expression (DEG) Analysis: Identifying genes that are uniquely expressed in specific cell clusters.
  • Ingenuity Pathway Analysis (IPA) and Gene Ontology (GO) Enrichment: To understand the biological functions and pathways associated with different cell subtypes.
  • Pseudotime Analysis: To infer developmental trajectories and dynamic states of cells.
  • Copy Number Variation (CNV) Analysis: Using CopyKAT to identify genomic alterations in malignant cells.
  • Cell-Cell Interaction (CCI) Analysis: Using tools like CellPhoneDB and CellTalkDB to predict interactions between different cell types based on ligand-receptor pairs.
  • Survival Analysis: Correlating the expression of specific gene signatures with patient survival using TCGA datasets.
  • Immunofluorescence Staining and RNAscope: For validating key findings at the protein and RNA level in tissue samples.

Datasets

  • Primary GC and Metastatic Samples: 10 fresh human tissue samples from six GC patients were collected for scRNA-seq.
  • TCGA-STAD Dataset: Used for survival analysis of gene signatures associated with malignant epithelial cell subclusters (n=407 patients).
  • GSE62254 Dataset: A large-scale primary GC dataset used to validate the 20-gene signature for lymph node metastasis.
  • GSE84437 Dataset: Another large-scale primary GC cohort used for prognostic significance of the 20-gene signature.
  • Normal GC Tissue Samples: Three normal GC tissue samples were downloaded from a public repository for correlation analysis with the adjacent non-tumoral sample.

Key Insights and Ideas

  • Metastasis is driven by distinct malignant epithelial subtypes: The study identified four malignant epithelial subclusters (G0-G3) with unique characteristics. G0 cells are associated with invasion and angiogenesis, G1 with EMT and peritoneal dissemination, G2 with dormancy, and G3 with EMT-induced cancer stem cell (CSC) phenotypes.
  • Immune microenvironment is key to metastasis: The study highlights the significant heterogeneity within immune and stromal cells, revealing how they collectively create a pro-tumoral and immunosuppressive microenvironment that supports metastasis.
  • MAIT cells and T cell-like B cells play a role in immune suppression: Unlike previous findings, MAIT cells in metastatic sites showed diminished effector function and an immunosuppressive phenotype. Similarly, T cell-like B cells were found to be associated with inhibiting T and NK cell activity.
  • Endothelial cells promote angiogenesis and immune resistance: Specific endothelial cell clusters (E0 and E3) were found to promote angiogenesis and vasculogenesis, while others (E1) actively suppressed T-cell responses.
  • Inflammatory CAFs are linked to GC growth and invasion: The study identified inflammatory CAFs (iCAFs) that express genes promoting extracellular matrix degradation and tumour cell invasion, and are associated with worse patient survival.
  • A 20-gene signature predicts lymph node metastasis: A novel signature derived from lymph node-derived exhausted CD8+ T cells was identified and validated, showing promise as a biomarker for predicting lymph node metastasis and patient prognosis.
  • HLA-E-KLRC1/KLRC2 interaction is a potential therapeutic target: This interaction was predicted to be involved in immune suppression across multiple cell types, suggesting it as a novel target for GC therapy.

Key Results

  • Identification of four malignant epithelial subclusters (G0-G3): These subclusters exhibited distinct functional characteristics related to invasion, EMT, dormancy, and CSC phenotypes.
  • Survival prediction by malignant epithelial subclusters: High expression of genes associated with G0, G1, and G3 subclusters predicted worse overall survival in GC patients.
  • Characterization of immune cell heterogeneity: The study detailed the diverse roles of T cells (including exhausted CD8+ T cells, Tregs, MAIT cells), B cells (including T cell-like B cells), myeloid cells (macrophages, monocytes, neutrophils), and dendritic cells in the GC microenvironment.
  • Discovery of a 20-gene signature for lymph node metastasis: This signature, derived from exhausted CD8+ T cells, was validated and showed significant prognostic value in GC cohorts.
  • Identification of potential therapeutic targets: HLA-E-KLRC1/KLRC2 interaction and targeting MDSCs were highlighted as potential therapeutic strategies.

Limitations and Future Work

  • Small sample size for scRNA-seq: The authors acknowledge that the sample size for scRNA-seq analysis was small, although findings were validated in larger bulk RNA-seq cohorts.
  • In vitro validation of some findings: While some interactions were predicted computationally, further in vitro and in vivo functional studies are needed to fully elucidate their roles.
  • Further investigation of novel cell types: The precise roles of newly identified cell types like Paneth-like cells in GC require further investigation.
  • Somatic mutation analysis: The authors suggest that parallel profiling of genetic and transcriptional heterogeneity would provide valuable information on somatic mutation evolution and functionality in GC.
  • Clinical trials for novel targets: The study points to the potential of targeting HLA-E-KLRC1 signalling as a novel clinical therapeutic opportunity, suggesting the need for clinical trials.

Summary updated

Abstract

This study provided insights into heterogeneous microenvironment of GC primary tumours and organ-specific metastases and provide support for precise diagnosis and treatment.
BibTeX
@inproceedings{jiang2022revealing,
  author = {Haiping Jiang and Dingyi Yu and Penghui Yang and Rongfang Guo and Mei Kong and Yuan Gao and Xiongfei Yu and Xiaoyan Lu and Xiaohui Fan},
  title = {{Revealing the transcriptional heterogeneity of organ‐specific metastasis in human gastric cancer using single‐cell RNA Sequencing}},
  booktitle = {Clinical},
  year = {2022},
  doi = {10.1002/ctm2.730},
}
Figure