Use of Single Cell Transcriptomic Techniques to Study the Role of High-Risk Human Papillomavirus Infection in Cervical Cancer

Abstract

High-risk human papillomavirus (hrHPV) infection has been associated with a higher probability of progression to cervical cancer. However, several extensive studies have reported that the presence of hrHPV can lead to a better prognosis, but the mechanism of how this occurs is unclear. In this study, microbiological analysis was used to identify HPV infection as a factor for the prognosis of patients with cervical squamous cell carcinoma (CSCC). Comparing the interactions of HPV⁺ and HPV^- malignant cells with immune cells as well as the trajectory of malignant cells either with or without HPV, we found that most of the HPV⁺ cells are well differentiated while HPV^- cells appear to be hypo-fractionated. Using transcriptomic and immunostaining data, we validated a set of unfavourable molecules in the HPV^- CSCC cells, including KRT16, ITGB1, CXCR1, VEGFA, CRCT1 and TNFRSF10B/DR5. This study provides a basis for the development of a rational post-operative follow-up programme and the development of an appropriate treatment plan for patients with cervical cancer.

Keywords: cervical cancer; high-risk human papillomavirus; prognosis markers; single cell transcriptomic; tumor infiltrating macrophages.

Figure 1

Virus analysis of the prognosis of high and low risk groups of CSCC patients. (A) Kaplan-Meier analysis of the risk groups that were defined with prognosis-correlated viruses in the TCGA dataset for CSCC patients. (B) Three-, five- and ten-year ROC survival curves of the risk groups for the CSCC TCGA dataset. (C) The contribution to the classification is ranked and alpha HPV is observed to support the low-risk group. Linear discriminant analysis was used to study the degree of influence of differential bacteria on sample groupings. (D) The abundance of alpha HPV in the low and high risk groups. (E) Kaplan-Meier survival analysis of the HPV⁺ group in CSCC patients.

Figure 2

Transcriptome profiling of HPV⁺ cells. (A) A UMAP diagram showing the cell clusters (B) UMAP diagrams showing HPV⁺ and - cells. (C) The bar graph quantifies and compares the proportion of HPV⁺ and ^- cells in the main cell types. (D) A bar graph to quantify and compare the proportions of HPV⁺ and HPV^- cells. (E) Violin plots indicating the expression of HPV in the main cell types. (F) The bar graph depicts the proportion of HPV high and low cells in the infected cell types. HPV high refers to the cells with hpv16 counts of more than 1.

Figure 3

CNV analyses of HPV⁺ tumor cells. (A) CNV plots showing the malignant cells in the normal and tumor tissues. (B) UMAP diagrams showing non- malignant and malignant cells in the normal and tumor tissues. (C) CNV scores in the tumor cell clusters. (D) Violin plots indicating the levels of malignant-related gene expression in the non- malignant and malignant cells. (E) Representative immunostained photomicrographs of KRT16 in the tissues from HPV^- and HPV⁺ patients. EpCam staining refers to the squamous cell carcinoma. Scale bar indicates 50μm. The bar graphs show the mean fluorescent intensity of the indicated staining. * indicates p<0.05 of six samples.

Figure 4

Trajectory analysis of HPV⁺ cervical cancer cells. (A) Pseudotime of cervical cancer cells. (B) Trajectory analysis of tumor cells in the seven states and in the correlated Seurat clusters (0, 2, 3 and 8). (C) The bar chart indicates the ratio of HPV^- and HPV⁺ tumor cells in different trajectory states. (D) Violin plots showing the expression of HPV16 in the seven states. (E) Volcano plots demonstrating the expression patterns and levels of the genes in State 1. (F) The enriched KEGG signaling pathway of GSEA analyses for HPV⁺ State 1 cell-correlated gene signatures. (G) The enriched KEGG signaling pathway of GSEA analyses for cancer stemness (State 1)-correlated gene signatures. (H) CNV scores in different trajectory states.

Figure 5

Cell-cell communication between immune cells and HPV⁺ tumor cells. (A) A network of interactions between immune and either HPV⁺ or HPV^- tumor cells. The lines represent the pointing relationships. (B) A Venn diagram showing the shared and unique cell-cell interactions of immune cells with HPV⁺ or HPV^- tumor cells. (C) Dot plots showing the most significant interactions (mean>1) of macrophages with either HPV⁺ or HPV^- tumor cells and the significance of their relationships. The horizontal coordinates are cell type interactions and the vertical coordinates are protein interactions, with the larger dots indicating smaller p-values and the colours representing the average expression. (D) Kaplan-Meier survival analysis of VEGFA, CD2, TNFRSF10B and TNFRSF1B in the CSCC patients. Representative immunostained photomicrographs of TNFRSF10B/DR5 (E) on the biopsied sections from HPV⁺ or HPV^- patients. CD206 and CD163 staining refer to the tumor infiltrated macrophages. Scale bar indicates 15μm. (F) Bar graphs showing the quantification of TNFRSF10B DR5 and CD206 respectively. ** indicate p<0.01 of six samples.

Figure 6

Workflow diagram of the current study.