Single-cell transcriptomic sequencing identifies subcutaneous patient-derived xenograft recapitulated medulloblastoma

Abstract

Background: Medulloblastoma (MB) is one of the most common malignant brain tumors that mainly affect children. Various approaches have been used to model MB to facilitate investigating tumorigenesis. This study aims to compare the recapitulation of MB between subcutaneous patient-derived xenograft (sPDX), intracranial patient-derived xenograft (iPDX), and genetically engineered mouse models (GEMM) at the single-cell level.

Methods: We obtained primary human sonic hedgehog (SHH) and group 3 (G3) MB samples from six patients. For each patient specimen, we developed two sPDX and iPDX models, respectively. Three Patch+/- GEMM models were also included for sequencing. Single-cell RNA sequencing was performed to compare gene expression profiles, cellular composition, and functional pathway enrichment. Bulk RNA-seq deconvolution was performed to compare cellular composition across models and human samples.

Results: Our results showed that the sPDX tumor model demonstrated the highest correlation to the overall transcriptomic profiles of primary human tumors at the single-cell level within the SHH and G3 subgroups, followed by the GEMM model and iPDX. The GEMM tumor model was able to recapitulate all subpopulations of tumor microenvironment (TME) cells that can be clustered in human SHH tumors, including a higher proportion of tumor-associated astrocytes and immune cells, and an additional cluster of vascular endothelia when compared to human SHH tumors.

Conclusions: This study was the first to compare experimental models for MB at the single-cell level, providing value insights into model selection for different research purposes. sPDX and iPDX are suitable for drug testing and personalized therapy screenings, whereas GEMM models are valuable for investigating the interaction between tumor and TME cells.

Keywords: experimental models; medulloblastoma; sPDX; single‐cell sequencing.

FIGURE 1

SHH and Group 3 MB tumor microenvironment compositions and characteristics. (A) Schematic overview of sample collection and model generation; (B, C) UMAP visualization of malignant and non‐malignant cell types of human SHH‐MB (n = 6) and G3‐MB (n = 6); (D, E) UMAP projection of representative marker genes in human SHH‐MB and G3‐MB; (F, G) The dotplot showing differentially expressed genes (DEGs) across the cell types in human SHH‐MB and G3‐MB; (H, I) GSVA analysis showing the activated pathway across different clusters of human SHH‐MB and human G3‐MB.

FIGURE 2

The subcutaneous PDX model better recapitulates SHH‐MB. (A, B) UMAP visualization of clusters identified in SHH‐MB intracranial PDX (iPDX) and subcutaneous PDX (sPDX) model; (C, D) UMAP projection of the marker genes in SHH‐MB iPDX and sPDX model; (E) GSVA analysis showing the activated pathway across different clusters of iPDX SHH‐MB models; (F) UMAP visualization of cells re‐clustered by primary human SHH‐MB, iPDX and sPDX SHH‐MB models.

FIGURE 3

Single‐cell Profiling of iPDX and sPDX model of G3‐MB. (A, B) UMAP visualization of cell types of G3‐MB iPDX and sPDX model; (C, D) UMAP projection of the marker genes in G3‐MB iPDX and sPDX model; (E, F) GSVA analysis showing the activated pathway across different clusters of G3‐MB iPDX and sPDX models; (G) UMAP visualization of cells re‐clustered by primary human G3‐MB, G3‐MB iPDX and sPDX models.

FIGURE 4

Single‐cell Profiling of SHH‐MB Genetically‐engineered mouse model. (A, B) UMAP visualization of cell types of SHH‐MB orthotopic GEMM (oGEMM) and subcutaneous GEMM (sGEMM); (C, D) GSVA analysis showing the activated pathway across different clusters of SHH‐MB oGEMM and sGEMM; (E, F) UMAP projection of the marker genes in oGEMM and sGEMM of SHH‐MB (G) UMAP visualization of cells re‐clustered by primary human SHH‐MB, oGEMM and sGEMM SHH‐MB models.

FIGURE 5

Multi‐model comparison of single‐cell transcriptomes across multiple models. (A) The proportion of different type of cells in SHH‐MB human sample and models; (B) The proportion of different type of cells in G3‐MB human sample and models; (C, D) Hierarchical clustering (Ward's method) of the average transcriptomes of high‐variation for snRNA‐sea data of each SHH‐MB model and human SHH‐MB, including all cells and including tumor cells only; (E, F) Hierarchical clustering (Ward's method) of the average transcriptomes of high‐variation for snRNA‐sea data of each G3‐MB model and human G3‐MB, including all cells and including tumor cells only; (G, H) Distributions of the Spearman correlation between each SHH‐MB model and bulk SHH‐MB using the expression signatures of all gene signatures in the SHH‐MB bulk data, including all cells and including tumor cells only; (I, J) Distributions of the Spearman correlation between each G3‐MB model and bulk G3‐MB using the expression signatures of all gene signatures in the G3‐MB bulk data, including all cells and including tumor cells only. HuMB, human medulloblastoma; iPDX, intracranial PDX model; sPDX, subcutaneous PDX model; Orth, orthotopic GEMM; Sub, subcutaneous GEMM.

FIGURE 6

Comparison of paired primary and recurrent SHH MB at single‐cell level. (A, B) UMAP visualization of human primary SHH‐MB and matched human recurrent SHH‐MB; (C, D) Heatmap showing the top10 DEGs across all the cell types in the primary SHH‐MB and in paired recurrent subgroup; (E) Predicted fraction of cells in human matched primary and recurrent SHH‐MB samples (n = 3) using RNA‐seq deconvolution.