Heterogeneity of human prostate carcinoma-associated fibroblasts implicates a role for subpopulations in myeloid cell recruitment.

Background: Carcinoma-associated fibroblasts (CAF) are a heterogeneous group of cells within the tumor microenvironment (TME) that can promote tumorigenesis in the prostate. By understanding the mechanism(s) by which CAF contributes to tumor growth, new therapeutic targets for the management of this disease may be identified. These studies determined whether unique sub-populations of human prostate CAF can be identified and functionally characterized.

Methods: Single-cell RNA-seq of primary human prostate CAF followed by unsupervised clustering was utilized to generate cell clusters based on differentially expressed (DE) gene profiles. Potential communication between CAF and immune cells was analyzed using in vivo tissue recombination by combining CAF or normal prostate fibroblasts (NPF) with non-tumorigenic, initiated prostate epithelial BPH-1 cells. Resultant grafts were assessed for inflammatory cell recruitment.

Results: Clustering of 3321 CAF allows for visualization of six subpopulations, demonstrating heterogeneity within CAF. Sub-renal capsule recombination assays show that the presence of CAF significantly increases myeloid cell recruitment to resultant tumors. This is supported by significantly increased expression of chemotactic chemokines CCL2 and CXCL12 in large clusters compared to other subpopulations. Bayesian analysis topologies also support differential communication signals between chemokine-related genes of individual clusters. Migration of THP-1 monocyte cells in vitro is stimulated in the presence of CAF conditioned medium (CM) compared with NPF CM. Further in vitro analyses suggest that CAF-derived chemokine CCL2 may be responsible for CAF-stimulated migration of THP-1 cells, since neutralization of this chemokine abrogates migration capacity.

Conclusions: CAF clustering based on DE gene expression supports the concept that clusters have unique functions within the TME, including a role in immune/inflammatory cell recruitment. These data suggest that CCL2 produced by CAF may be involved in the recruitment of inflammatory cells, but may also directly regulate the growth of the tumor. Further studies aimed at characterizing the subpopulation(s) of CAF which promote immune cell recruitment to the TME and/or stimulate prostate cancer growth and progression will be pursued.