Panoply™ Human CSF1 Knockdown Stable Cell Line

Pancreatic ductal adenocarcinoma (PDAC) is characterized by an abundant stroma. Molecular subtyping studies have identified a mesenchymal tumor subtype consistently associated with high-grade malignancy and poor clinical outcomes. However, the relationship between the PDAC stroma and tumor subtypes remains unclear. Here, researchers investigated how PDAC cells regulate pancreatic stellate cells (PSCs), the primary cellular component of the stroma. Analysis of primary tissues revealed that high-grade PDAC exhibits reduced collagen deposition compared to low-grade PDAC. Xenograft models and organotypic co-culture systems established using mesenchymal-like PDAC cells demonstrated lower collagen content and reduced levels of activated PSCs. Conditioned medium transfer experiments involving a large panel of PDAC cell lines showed that mesenchymal-like PDAC cells consistently downregulate ACTA2 and COL1A1 expression in PSCs and inhibit their proliferation. The researchers identified colony-stimulating factor 1 (CSF-1) as the mesenchymal-PDAC-derived ligand responsible for PSC inactivation, noting that inhibiting its receptor, CSF1R, reversed this effect. In summary, the stromal characteristics of high-grade PDAC include low collagen content and low levels of activated PSCs; targeting CSF1R offers a direct strategy for maintaining a tumor-suppressive microenvironment.

To investigate whether the inactivation of PSCs induced by mesenchymal-like PDAC cells could be prevented, the researchers conducted CSF-1 inhibition experiments. Compared with the control group, selective inhibition of CSF1R (using anti-CSF1R) did not alter the morphology of PS-1 cells exposed to PS-1 conditioned medium (CM), nor did it affect the transcriptional levels of ACTA2 and COL1A1 (Figure 1A and B). In contrast, when PS-1 cells were exposed to CM from mesenchymal-like cells containing anti-CSF-1 antibodies, the aforementioned inactivation effect was partially reversed, evidenced by attenuated morphological changes and increased expression of stromal activation markers (Figure 1A and B). The researchers validated these inhibition results using CSF-1 knockdown PANC-1 cells. PS-1 cells exposed to CM from PANC-1 cells transduced with a control scrambled sequence vector (pLKOctrl) exhibited an elongated morphology similar to that observed upon exposure to CM from untransduced PANC-1 cells (Figure 1C). However, these morphological effects were attenuated in the presence of CSF-1 knockdown PANC-1 cells (Figure 1C). The degree of CSF-1 knockdown in PANC-1 cells correlated with COL1A1 expression levels in PS-1 cells (Figure 1D); specifically, CM from PANC-1 cells with the lowest CSF-1 expression induced the highest levels of COL1A1 expression in PS-1 cells (Figure 1D). Given that complete knockdown of CSF-1 was not achieved after screening, this suggests that CSF-1 secretion is an intrinsic characteristic of mesenchymal-like PDAC cells. To further confirm that tumor-derived CSF-1 drives PSC inactivation, the researchers overexpressed CSF-1 in Capan-2 cells using a lentiviral vector (Figure 1F). The results showed that, compared with CM from control vector-transfected Capan-2 cells, CM from Capan-2 cells secreting high levels of CSF-1 induced PSC inactivation, as evidenced by changes in cell morphology (Figure 1E) and reduced expression levels of COL1A1 and ACTA2 (Figure 1G). These gene silencing and overexpression experiments collectively demonstrate that CSF-1 produced by tumor cells leads to PSC inactivation, and that targeting CSF-1 can block this process.

Figure 1. Targeting CSF‐1 in mesenchymal‐like PDAC cells abolishes PSC deactivation. (Steins A, et al., 2020)