Hepatic Stellate Cells Alter Liver Immune Environment to Promote Cancer
Activation of hepatic stellate cells is associated with an increased expression of M2-related genes, suggesting monocyte reprogramming to an immunosuppressive state.
Hepatocellular carcinoma (HCC) is the most common form of liver cancer, accounting for up to 90 percent of cases, and is the second most common cause of cancer-related deaths worldwide according to the World Health Organization’s 2014 World Cancer Report. Even when caught early, HCC often recurs, either from intra-liver metastases or new primary tumors, and recurrence is the leading cause of death for patients with HCC. The liver microenvironment is an important contributor to HCC initiation and progression and also likely plays a role in tumor recurrence. Xin Wei Wang, Ph.D., of CCR’s Laboratory of Human Carcinogenesis, and his colleagues wondered whether activated hepatic stellate cells (A-HSCs), stromal cells in the liver known to participate in repair following injury and in the development of fibrosis, contribute directly to HCC recurrence.
The researchers began their studies by evaluating the gene expression of human primary hepatocytes, primary fibroblasts, and primary cultured A-HSCs to establish an A-HSC-specific gene signature. They then performed survival risk prediction based on the expression of A-HSC genes in tissue surrounding HCCs of patients in a large cohort study. The A-HSC-specific gene set assigned cases into high- and low-risk groups, which correlated with overall survival. Using the gene signature to assign risk in matched tumor tissue, however, did not associate with survival. In two additional cohorts the expression of A-HSC related genes in tissue near the tumor predicted tumor recurrence; in a fourth cohort the gene signature predicted patient survival. Univariate and multivariate Cox proportional hazards regression analysis showed that the gene signature was a significant and independent predictor of survival. By comparing global gene expression profiles of HCC tissues from high- and low-risk patients determined by the gene signature, the investigators found no significant differences. Together these results suggest that A-HSCs contribute to a microenvironment that favors HCC progression.
Further examination of the A-HSC gene signature revealed 37 genes that significantly correlated with patient survival. Clustering based on these genes separated samples into two main groups, one enriched for A-HSC high-risk cases and one enriched for A-HSC low-risk cases, and the scientists used quantitative reverse transcriptase polymerase chain reaction to validate the expression of three highly-upregulated genes, CCL2, SRGN, and JAG1. Pathway analysis revealed that most of the 37 genes affected functions such as HSC activation as well as epithelial-mesenchymal transition and growth factor signaling. When the researchers compared their signature to two previously published HCC stroma signatures, they found that the three were largely distinct, though the A-HSC signature and one of the others assigned cases to similar groups.
The investigators then compared gene expression in tissues surrounding HCCs from their initial cohort and found nearly 3,000 genes differentially expressed between high- and low-risk groups. Pathway analysis showed that most of the enriched pathways were related to inflammation or immune responses as well as HSC activation, suggesting inflammatory or immune-related activity may contribute to poor HCC prognosis associated with A-HSCs. The scientists performed transcriptome analysis of immune cell-specific genes and found enrichment of myeloid but not lymphoid lineages in high-risk cases. They validated this observation with immunohistochemistry, revealing that myeloid cells, including monocytes, are the main contributor of poor HCC prognosis associated with A-HSCs.
To understand how A-HSCs affect immune cells, the researcher grew peripheral blood mononuclear cells (PBMCs) from healthy donors in conditioned medium from the HSC cell line LX2. The expression of CD14 on monocytes increased, but other monocyte markers and T-cell markers showed no difference. Co-culturing CD14-positive PBMCs with LX2 cells further enhanced CD14 expression and increased expression of CD15 and CCR2. However, expression of the T-lymphocyte antigen CD86/B7-2 decreased in the presence of LX2 cells. Growing CD14-negative cells with LX2 cells did not increase T cell activation markers. These results show that HSCs affect only monocytes and that this effect was enhanced by co-culturing the cells.
The investigators next looked at the expression of monocyte-related genes in tissues near tumors. In high-risk samples, they found higher levels of M2-like genes while in low-risk samples M1-like genes predominated. Similarly, in CD14-positive PBMCs co-cultured with LX2 cells, seven M2-like genes were upregulated and two of seven M1-like genes were downregulated, suggesting A-HSCs may reprogram monocytes toward an immunosuppressive phenotype.
Lastly, the scientists studied HCC cell migration, proliferation, and spheroid formation in the presence of A-HSCs. Tumor cell migration was enhanced when the cells were cultured in conditioned medium from LX2 cells or co-cultured with LX2 cells. Migration further increased in the presence of CD14-positive, but not CD14-negative, cells. In contrast, HCC cell proliferation increased with LX2 cells alone but was reduced by including CD14-positive cells. Culturing tumor cells with CD14-positive cells alone did not increase proliferation. Sphere formation, a measure of cancer cell self-renewal, was promoted by culturing HCC cells with CD14-positive cells but not LX2 cells. However, culturing the HCC cells with both CD14-positive PBMCs and LX2 cells enhanced sphere formation.
Taken together these studies reveal that A-HSCs play an important role in promoting HCC progression by interacting with and altering the profile of monocytes in the liver microenvironment. Disrupting these interactions may be one effective way to prevent HCC recurrences.Summary Posted: 05/2015
Ji J, Eggert T, Budhu A, Forgues M, Takai A, Dang H, Ye Q, Lee JS, Kim JH, Greten TF, and Wang XW. Hepatic Stellate Cell and Monocyte Interaction Contributes to Poor Prognosis in Hepatocellular Carcinoma. Hepatology. April 1, 2015 PubMed Link