Exploring the impact of PDGFD in osteosarcoma metastasis through single-cell sequencing analysis

Abstract

Purpose: The overall survival rate for metastatic osteosarcoma hovers around 20%. Responses to second-line chemotherapy, targeted therapies, and immunotherapies have demonstrated limited efficacy in metastatic osteosarcoma. Our objective is to validate differentially expressed genes and signaling pathways between non-metastatic and metastatic osteosarcoma, employing single-cell RNA sequencing (scRNA-seq) and additional functional investigations. We aim to enhance comprehension of metastatic mechanisms and potentially unveil a therapeutic target.

Methods: scRNA-seq was performed on two primary osteosarcoma lesions (1 non-metastatic and 1 metastatic). Seurat package facilitated dimensionality reduction and cluster identification. Copy number variation (CNV) was predicted using InferCNV. CellChat characterized ligand-receptor-based intercellular communication networks. Differentially expressed genes underwent GO function enrichment analysis and GSEA. Validation was achieved through the GSE152048 dataset, which identified PDGFD-PDGFRB as a common ligand-receptor pair with significant contribution. Immunohistochemistry assessed PDGFD and PDGFRB expression, while multicolor immunofluorescence and flow cytometry provided insight into spatial relationships and the tumor immune microenvironment. Kaplan-Meier survival analysis compared metastasis-free survival and overall survival between high and low levels of PDGFD and PDGFRB. Manipulation of PDGFD expression in primary osteosarcoma cells examined invasion abilities and related markers.

Results: Ten clusters encompassing osteoblasts, osteoclasts, osteocytes, fibroblasts, pericytes, endothelial cells, myeloid cells, T cells, B cells, and proliferating cells were identified. Osteoblasts, osteoclasts, and osteocytes exhibited heightened CNV levels. Ligand-receptor-based communication networks exposed significant fibroblast crosstalk with other cell types, and the PDGF signaling pathway was activated in non-metastatic osteosarcoma primary lesion. These results were corroborated by the GSE152048 dataset, confirming the prominence of PDGFD-PDGFRB as a common ligand-receptor pair. Immunohistochemistry demonstrated considerably greater PDGFD expression in non-metastatic osteosarcoma tissues and organoids, correlating with extended metastasis-free and overall survival. PDGFRB expression showed no significant variation between non-metastatic and metastatic osteosarcoma, nor strong correlations with survival times. Multicolor immunofluorescence suggested co-localization of PDGFD with PDGFRB. Flow cytometry unveiled a highly immunosuppressive microenvironment in metastatic osteosarcoma. Manipulating PDGFD expression demonstrated altered invasive abilities and marker expressions in primary osteosarcoma cells from both non-metastatic and metastatic lesions.

Conclusions: scRNA-seq illuminated the activation of the PDGF signaling pathway in primary lesion of non-metastatic osteosarcoma. PDGFD displayed an inhibitory effect on osteosarcoma metastasis, likely through the suppression of the EMT signaling pathway.

Keywords: Metastasis; Organoid; Osteosarcoma; PDGFD; scRNA-seq.

Fig. 1

Cell profiling in osteosarcoma assessed by scRNA-seq. a Flow chart for the collection and processing of primary lesions from patients with osteosarcoma for scRNA-seq. b UMAP plot displayed ten identified 10 cell types. c UMAP plot based on samples. d Dot plots showed the signature gene expressions across the 10 cell types. e The relative proportions of cell types in each sample

Fig. 2

Cell communication network analysis. a Connectome networks in non-metastatic osteosarcoma. Dots indicate cell types. The thickness of line is proportional to the number of interactions between two cell types. The loops indicate autocrine circuits. b Connectome networks in metastatic osteosarcoma. c Heatmap of the outgoing signaling patterns. d Heatmap of the incoming signaling patterns

Fig. 3

The identification of the crucial PDGF-PDGFR pair. a Violin plots showed the expression of different subtypes of PDGF-PDGFR pair in discovery cohort dataset. b Dot plot depicted different subtypes of PDGF-PDGFR pair in discovery cohort dataset. The dot color and size represent the calculated communication probability and p-values. c Relative contribution of each PDGF-PDGFR pair in discovery cohort dataset. d Violin plots showed the expression of different subtypes of PDGF-PDGFR pair in GSE152048 dataset. e Dot plot depicted different subtypes of PDGF-PDGFR pair in GSE152048 dataset. f Relative contribution of each PDGF-PDGFR pair in GSE152048 dataset

Fig. 4

The expression of PDGFD and PDGFRB in osteosarcoma. a UMAP plot displayed PDGFD highly expressed in the non-metastatic osteosarcoma. b UMAP plot displayed similar expression of PDGFRB in the non-metastatic and metastatic osteosarcoma. c Immunohistochemical staining for PDGFD and PDGFRB in samples and organoids. Black scale bars: 30 μm. d PDGFD positivity was significantly higher in non-metastatic osteosarcoma primary lesions than metastatic primary lesions. e The differing positive rate of PDGFRB between non-metastatic osteosarcoma primary lesions and metastatic primary lesions was not statistically significant. * p < 0.05. ns, not significant, p > 0.05. f Multi-color immunofluorescence analysis revealed that PDGFD and PDGFRB were co-localized. Color code: Blue = DAPI, Red = COL1a1, Green = PDGFD, Yellow = PDGFRB. White scale bars: 40 μm.

Fig. 5

Kaplan-Meier survival curves of metastasis-free survival and overall survival. a Metastasis-free survival was significantly longer in the PDGFD-high group than in the PDGFD-low group. b Overall survival was significantly longer in the PDGFD-high group than in the PDGFD-low group. c Metastasis-free survival was not significantly different between PDGFRB-high group and the PDGFRB-low group. d Overall survival was not significantly different between PDGFRB-high group and the PDGFRB-low group.

Fig. 6

The highly immunosuppressive microenvironment in metastatic osteosarcoma. a The gating strategy of multicolor flow cytometry. b Bar chart of the percentage of immune cell subsets in primary lesions of osteosarcoma. c Bar chart of the percentage of immune cell subsets in peripheral blood mononuclear cells of osteosarcoma

Fig. 7

The relationship between PDGFD-PDGFRB and tumor immune microenvironment in osteosarcoma. Color code: Blue = DAPI, Green = PDGFD, Yellow = PDGFRB, White = CD4 or CD86, Red = CD8 or CD206. White scale bars: 40 μm

Fig. 8

PDGFD inhibit the metastasis ability of osteosarcoma. a Invasion assays for primary osteosarcoma cells derived from non-metastatic osteosarcoma with PDGFD overexpression or knockdown. b Invasion assays for primary osteosarcoma cells derived from metastatic osteosarcoma with PDGFD overexpression or knockdown. c-d bar graph showed the number of invade cells in primary osteosarcoma cells derived from non-metastatic osteosarcoma with PDGFD overexpression or knockdown. e-f bar graph showed the number of invade cells in primary osteosarcoma cells derived from metastatic osteosarcoma with PDGFD overexpression or knockdown. g-h Bar chart of fold change of GAPDH-normalized mRNA levels of indicated genes in primary osteosarcoma cells derived from non-metastatic osteosarcoma with PDGFD overexpression or knockdown. i-j Bar chart of fold change of GAPDH-normalized mRNA levels of indicated genes in primary osteosarcoma cells derived from metastatic osteosarcoma with PDGFD overexpression or knockdown. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001