Increased Stiffness Downregulates Focal Adhesion Kinase Expression in Pancreatic Cancer Cells Cultured in 3D Self-Assembling Peptide Scaffolds

Abstract

The focal adhesion kinase (FAK) is a non-receptor tyrosine kinase that participates in integrin-mediated signal transduction and contributes to different biological processes, such as cell migration, survival, proliferation and angiogenesis. Moreover, FAK can be activated by autophosphorylation at position Y397 and trigger different signaling pathways in response to increased extracellular matrix stiffness. In addition, FAK is overexpressed and/or hyperactivated in many epithelial cancers, and its expression correlates with tumor malignancy and invasion potential. One of the characteristics of solid tumors is an over deposition of ECM components, which generates a stiff microenvironment that promotes, among other features, sustained cell proliferation and survival. Researchers are, therefore, increasingly developing cell culture models to mimic the increased stiffness associated with these kinds of tumors. In the present work, we have developed a new 3D in vitro model to study the effect of matrix stiffness in pancreatic ductal adenocarcinoma (PDAC) cells as this kind of tumor is characterized by a desmoplastic stroma and an increased stiffness compared to its normal counterpart. For that, we have used a synthetic self-assembling peptide nanofiber matrix, RAD16-I, which does not suffer a significant degradation in vitro, thus allowing to maintain the same local stiffness along culture time. We show that increased matrix stiffness in synthetic 3D RAD16-I gels, but not in collagen type I scaffolds, promotes FAK downregulation at a protein level in all the cell lines analyzed. Moreover, even though it has classically been described that stiff 3D matrices promote an increase in pFAK^Y397/FAK proteins, we found that this ratio in soft and stiff RAD16-I gels is cell-type-dependent. This study highlights how cell response to increased matrix stiffness greatly depends on the nature of the matrix used for 3D culture.

Keywords: FAK; PDAC; RAD16-I; biomechanics; pancreatic ductal adenocarcinoma; self-assembling peptides; stiffness.

Figure 1

Cell culture in 3D RAD16-I scaffolds. (a) Cell morphology in 2D and 3D cultures. Actin (Phalloidin, pseudo-colored in yellow) and nuclei (DAPI, blue) staining of PANC-1, MiaPaCa-2 and hNDF cells cultured in classic 2D dishes and RAD16-I 3D scaffolds at two different peptide concentrations. Scale bars represent 20 µm; (b) macroscopic view of 3D constructs stained with Congo red for clearer visualization; (c) growth curves of PANC-1 and MiaPaCa-2 cells in 0.15% and 0.5% RAD16-I.

Figure 2

Protein adsorbing into RAD16-I hydrogels. (a) BCA protein assay of RAD16-I gels. The graph shows the OD values obtained for known concentrations of protein (50 μg/mL−500 μg/mL BSA) and RAD16-I hydrogels assembled with 10% FBS or PBS; (b) fibronectin (green) immunofluorescence of PANC-1 and MiaPaCa-2 cells in RAD16-I 3D culture counterstained with phalloidin (red) and DAPI (blue). Scale bars represent 20 µm.

Figure 3

Immunofluorescence analysis of signal mechanotransduction proteins. Actin (blue) β1-integrin (green), pFAK^Y397 (red), pFAK^Y861 (red) and vinculin (red) staining in PANC-1 (left) and MiaPaCa-2 (right) cells in (a) 0.15% RAD16-I; (b) 0.5% RAD16-I and (c) 2D cultures. Merge of green and red channels is also shown. Scale bars represent 10 µm; (d) Mander’s colocalization coefficients of PANC-1 and MiaPaCa-2 cells (n = 5). Statistical differences are indicated as * for p value < 0.05, ** for p value < 0.01 and *** for p value < 0.001.

Figure 4

pFAK^Y397 and FAK expression in cancer cell lines and normal fibroblasts cultured in RAD16-I hydrogels. (a) Western blot bands of pFAK^Y397 and total FAK in PANC-1, MiaPaCa-2 and hNDF cells cultured in 2D and 0.15% (soft) and 0.5% (stiff) RAD16-I gels; (b) densitometry of bands shown in (a); (c) Western blot bands of pFAK^Y397 and total FAK in PANC-1 cells cultured in 3D RAD16-I hydrogels of increased stiffness (from 0.15% to 0.8% RAD16-I, representing a stiffness range of 100–8500 Pa); (d) densitometry of bands shown in (c). GAPDH was used as loading control. Statistical differences are indicated as * for p value < 0.05, ** for p value < 0.01, *** for p value < 0.001 and **** for p value < 0.0001.

Figure 5

Cell culture in RAD16-I matrix functionalized with collagen type I. (a) SDS-PAGE of RAD or RAD/COL cell-free gels; (b) phase-contrast microscopy images of fibroblast 24 h after cell embedding within 0.15% RAD or RAD/COL gels. Scale bars represent 50 µm; (c) Western blot bands of FAK and Erk1/2 and its phosphorylated forms in PANC-1 cells cultured in 0.15% (soft) and 0.5% (stiff) RAD16-I gels mixed with increasing amounts of collagen I; (d) densitometry of pFAK^Y397/FAK shown in (c). GAPDH was used as loading control. Statistical differences are indicated as ** for p value < 0.01, *** for p value < 0.001 and **** for p value < 0.0001.

Figure 6

Cell culture in 3D collagen type I gels. (a) Macroscopic view of collagen 3D constructs containing PANC-1 cells; (b) Western blot bands of pFAK^Y397 and total FAK in PANC-1, MiaPaCa-2 and MCF-7 cells cultured in 0.15% (soft) and 0.5% (stiff) collagen gels; (c) densitometry of bands shown in (b) represented as pFAK^Y397/FAK; (d) Western blot bands of pFAK^Y397 and total FAK in PANC-1 cells cultured in 0.15% (soft) and 0.5% (stiff) collagen gels in the presence of GM6001.