Integrated analysis of the rhesus monkey liver transcriptome during development and human primary HCC AFP-related gene expression

Abstract

Embryonic development and tumorigenesis have a certain degree of similarity. Alpha-fetoprotein (AFP), a protein related to embryonic development, is a well-known biomarker for the diagnosis and prognosis of hepatocellular carcinoma (HCC). In this study, we analyzed the differences in gene expression profiles and molecular mechanisms in human HCC tissues from patients in AFP^high (serum AFP level ≥ 25 ng/mL) and AFP^low (serum AFP level < 25 ng/mL) groups. The results indicated that AFP^high HCC has more malignant biological characteristics. Single-sample gene set enrichment analysis (ssGSEA) showed significantly higher levels of genes expressed in dendritic cells, neutrophils, and natural killer cells in the AFP^low group than in the AFP^high group. Then, we defined a rhesus monkey fetal liver developmental landscape and compared it to the HCC gene expression profile. The gene signatures of AFP^high HCC tissues were similar to those of early embryonic liver tissues. In this study, we comprehensively analyzed the rhesus monkey liver transcriptome during development and human primary HCC AFP-related gene expression profiles and clarified the function of AFP in the occurrence and development of HCC from the perspective of developmental biology, which might provide a new perspective on the pathogenesis of HCC.

Keywords: AFP; embryonic liver development; hepatocellular carcinoma; rhesus monkey; tumor immunity.

Graphical abstract

Figure 1

Analysis of the prognostic value of AFP expression levels in patients with HCC (A and B) Kaplan-Meier plots of DFS/OS times for patients with HCC stratified by AFP expression levels. Notably, 25 ng/mL served as the cutoff value according to the clinical application.

Figure 2

Characteristics of gene expression in patients with AFP^high and AFP^low HCC (A) The t-SNE analysis shows a significant distinction between patients with high and low AFP levels. (B) Volcano plot of the differentially expressed genes between the two groups of patients. (C and D) KEGG pathway analysis of upregulated and downregulated genes in AFP^high HCC. (E–G) KEGG pathway-related gene set identified by the GSEA algorithm in AFP^low HCC samples.

Figure 3

Expression levels of representative genes involved in the identified differential altered pathways between patients with AFP^high and AFP^low HCC (A) Expression levels of representative genes involved in cell cycle processes (CDK1, CDC25C, CCNB1, and CCNB2) and metabolic processes (CYP2A13, UGT2B15, and HSD11B1) in the real-time PCR cohort. (B) Expression levels of the aforementioned genes in the cohort TCGA-LIHC. Statistical method: rank sum test.

Figure 4

Comparison of related immune cells between AFP^high and AFP^low HCC by scoring immune cell-specific genes (A) Dendritic cells. (B) Neutrophils. (C) Natural killer cells. Statistic method: rank sum test.

Figure 5

The developmental landscape of the embryonic liver in rhesus macaque (A) The distribution of all homologous genes identified using PCA. (B) The distribution of T1–T4 developmental samples on the landscape. (C) Homologous genes were divided into 9 groups. (D) Overall trend of gene expression in each group.

Figure 6

HCC samples were projected onto the developmental landscape (A and B) The average distance between AFP^high and AFP^low HCC samples with tissues from each developmental stage. (C and D) The distribution of differentially expressed genes between AFP^high and AFP^low HCC in the gene development landscape.