PD-173074

Breast cancer progression involves intricate interactions between cancer cells and the tumor microenvironment (TME). This study elucidates the critical role of progesterone receptor (PR) signaling in mediating the protumorigenic effects of cancer-associated fibroblasts (CAFs) on estrogen receptor–positive (ER+) luminal breast cancer cells. We demonstrate that CAFs produce physiologically relevant levels of estrogen and progesterone, which significantly contribute to breast cancer tumorigenicity. Specifically, CAF conditioned media (CM) promoted PR-dependent anchorage-independent growth, tumorsphere formation/stem cell expansion, and CD44 upregulation. CAF cells formed co-clusters more frequently with PR+ breast cancer cells relative to PR-null models. While both PR isoforms mediated these actions, PR-A was a dominant driver of tumorsphere formation/stemness, while PR-B induced robust CD44 expression and CAF/tumor cell co-cluster formation. CD44 knockdown impaired CAF/tumor cell co-clustering. Fibroblast growth factor 2 (FGF2), also secreted by CAFs, phosphorylated PR (Ser294) in a MAPK-dependent manner and activated PR to enhance CD44 expression and breast cancer tumorigenicity. The FGF receptor (FGFR) inhibitor PD173074 diminished CAF– and FGF2–dependent PR activation, tumorsphere formation, and co-clustering. In summary, this study reveals a novel mechanism through which stromal CAFs orchestrate elevated PR signaling in ER+ luminal breast cancer via secretion of both progesterone and FGF2, a potent activator of ERK1/2. Understanding tumor cell/TME interactions provides insights into potential therapeutic strategies aimed at disrupting PR– and/or FGF2/FGFR-dependent signaling pathways to prevent early metastasis in patients with ER+ breast cancer.

Large, multinucleated cells, like syncytiotrophoblasts (STB), are not readily analyzed by standard methods used for single cells, such as single-cell RNA-sequencing and fluorescence-activated cellular sorting (FACS). Here we have demonstrated that fluorescence-activated nuclear sorting (FANS) is suitable to analyze nuclei from STB. Human pluripotent stem cells (PSCs) can be differentiated into a mixed trophoblast populations comprising approximately 20 % STB by treatment with BMP4 (Bone Morphogenetic Protein-4), plus A83-01 and PD173074, inhibitors of activin and FGF2 signaling, respectively (the BAP model) in about a week. Here we demonstrate that FANS can be used to separate two types of STB nuclei from the nine different clusters of trophoblast nuclei previously identified in the BAP model by single nucleus RNA sequencing (snRNAseq). Rather than using cell surface markers, as in FACS, transcription factors in various combinations were employed to target specific nuclear types. Nuclei were isolated at d 8 of BAP differentiation of H1 human embryonic stem cells and fixed in 4 % paraformaldehyde. After permeabilization in 0.1 % triton X-100, nuclei were incubated for 3 and 1 h at 4 °C with primary and secondary antibodies respectively and nuclear samples were then subjected to FANS. By using markers identified by snRNA and immunohistochemistry, nuclei were first sorted into a Topoisomerase-1, or TOP1, bright population and then into the two STB subpopulations by using antibodies to JUNB (Jun B Proto-Oncogene) and TFCP2L1 (Transcription Factor CP2 Like 1). The protocol established here is simple, straightforward, and efficient and can be used on a relatively large scale to sort individual subtypes of nuclei from mixed populations of trophoblasts for further analysis.