Cell mediated remodeling of stiffness matched collagen and fibrin scaffolds.

Cells are known to continuously remodel their local extracellular matrix (ECM) and in a reciprocal way, they can also respond to mechanical and biochemical properties of their fibrous environment. In this study, we measured how stiffness around dermal fibroblasts (DFs) and human fibrosarcoma HT1080 cells differs with concentration of rat tail type 1 collagen (T1C) and type of ECM. Peri-cellular stiffness was probed in four directions using multi-axes optical tweezers active microrheology (AMR). First, we found that neither cell type significantly altered local stiffness landscape at different concentrations of T1C. Next, rat tail T1C, bovine skin T1C and fibrin cell-free hydrogels were polymerized at concentrations formulated to match median stiffness value. Each of these hydrogels exhibited distinct fiber architecture. Stiffness landscape and fibronectin secretion, but not nuclear/cytoplasmic YAP ratio differed with ECM type. Further, cell response to Y27632 or BB94 treatments, inhibiting cell contractility and activity of matrix metalloproteinases, respectively, was also dependent on ECM type. Given differential effect of tested ECMs on peri-cellular stiffness landscape, treatment effect and cell properties, this study underscores the need for peri-cellular and not bulk stiffness measurements in studies on cellular mechanotransduction.

Patterned photocrosslinking to establish stiffness anisotropies in fibrous 3D hydrogels.

Cells are known to constantly interact with their local extracellular matrix (ECM) and respond to a variety of biochemical and mechanical cues received from the ECM. Nonetheless, comprehensive understanding of cell-ECM interactions has been elusive. Many studies rely on analysis of cell behavior on 2D substrates, which do not reflect a natural cell environment. Further, lack of dynamic control over local stiffness anisotropies and fiber alignment hinders progress in studies in naturally derived fibrous 3D cultures. Here, we present a cell-safe method of patterned photocrosslinking, which can aid in studying biological hypotheses related to mechanotransduction in 3D hydrogels. As previously described by our group, ruthenium-catalyzed photocrosslinking (RCP) of selected ECM regions promotes localized increase in stiffness mediated by focused blue laser light in a confocal microscope. In this study, we further demonstrate that RCP can induce localized strain stiffening and fiber alignment outside of the selected crosslinked region and induce stiffness anisotropy biased towards the direction of fiber alignment. MDA-MB-231 cells are shown to respond to RCP-induced changes in local ECM architecture and display directional bias towards the direction of fiber alignment, as compared to control cells. Further, the effect of patterned crosslinking on a stiffness landscape is measured using multi-axes optical tweezers active microrheology (AMR) with backscattered laser beam illumination. AMR validates RCP as a suitable tool for creating distinct stiffness anisotropies which promote directed migration of cells, further underscoring the usefulness of RCP in cell-ECM studies. STATEMENT OF SIGNIFICANCE: Studies on cell-ECM interactions in 3D cultures have often been hindered by the lack of available tools to dynamically alter local ECM stiffness and fiber alignment. Here, we present a non-invasive, cell-safe and easily applicable method of patterned photocrosslinking, which can aid in studying biological hypotheses in fibrous 3D hydrogels. Ruthenium-catalyzed crosslinking (RCP) of selected fibrin ECM regions promotes localized increase in stiffness and creates distinct stiffness anisotropies in the presence of the focused blue laser light. Outside of the crosslinked region, RCP causes fiber alignment and strain stiffening in the ECM, verified using multi-axes optical tweezers active microrheology (AMR). Following RCP, human breast cancer MDA-MB-231 exhibit directed cell migration, validating usefulness of this method in cell-ECM studies.

Cell contact guidance via sensing anisotropy of network mechanical resistance.

Despite the ubiquitous importance of cell contact guidance, the signal-inducing contact guidance of mammalian cells in an aligned fibril network has defied elucidation. This is due to multiple interdependent signals that an aligned fibril network presents to cells, including, at least, anisotropy of adhesion, porosity, and mechanical resistance. By forming aligned fibrin gels with the same alignment strength, but cross-linked to different extents, the anisotropic mechanical resistance hypothesis of contact guidance was tested for human dermal fibroblasts. The cross-linking was shown to increase the mechanical resistance anisotropy, without detectable change in network microstructure and without change in cell adhesion to the cross-linked fibrin gel. This methodology thus isolated anisotropic mechanical resistance as a variable for fixed anisotropy of adhesion and porosity. The mechanical resistance anisotropy |Y*| -1 - |X*| -1 increased over fourfold in terms of the Fourier magnitudes of microbead displacement |X*| and |Y*| at the drive frequency with respect to alignment direction Y obtained by optical forces in active microrheology. Cells were found to exhibit stronger contact guidance in the cross-linked gels possessing greater mechanical resistance anisotropy: the cell anisotropy index based on the tensor of cell orientation, which has a range 0 to 1, increased by 18% with the fourfold increase in mechanical resistance anisotropy. We also show that modulation of adhesion via function-blocking antibodies can modulate the guidance response, suggesting a concomitant role of cell adhesion. These results indicate that fibroblasts can exhibit contact guidance in aligned fibril networks by sensing anisotropy of network mechanical resistance.

Dermal fibroblasts and triple-negative mammary epithelial cancer cells differentially stiffen their local matrix.

The bulk measurement of extracellular matrix (ECM) stiffness is commonly used in mechanobiology. However, past studies by our group show that peri-cellular stiffness is quite heterogeneous and divergent from the bulk. We use optical tweezers active microrheology (AMR) to quantify how two phenotypically distinct migratory cell lines establish dissimilar patterns of peri-cellular stiffness. Dermal fibroblasts (DFs) and triple-negative human breast cancer cells MDA-MB-231 (MDAs) were embedded within type 1 collagen (T1C) hydrogels polymerized at two concentrations: 1.0 mg/ml and 1.5 mg/ml. We found DFs increase the local stiffness of 1.0 mg/ml T1C hydrogels but, surprisingly, do not alter the stiffness of 1.5 mg/ml T1C hydrogels. In contrast, MDAs predominantly do not stiffen T1C hydrogels as compared to cell-free controls. The results suggest that MDAs adapt to the bulk ECM stiffness, while DFs regulate local stiffness to levels they intrinsically prefer. In other experiments, cells were treated with transforming growth factor-ß1 (TGF-ß1), glucose, or ROCK inhibitor Y27632, which have known effects on DFs and MDAs related to migration, proliferation, and contractility. The results show that TGF-ß1 alters stiffness anisotropy, while glucose increases stiffness magnitude around DFs but not MDAs and Y27632 treatment inhibits cell-mediated stiffening. Both cell lines exhibit an elongated morphology and local stiffness anisotropy, where the stiffer axis depends on the cell line, T1C concentration, and treatment. In summary, our findings demonstrate that AMR reveals otherwise masked mechanical properties such as spatial gradients and anisotropy, which are known to affect cell behavior at the macro-scale. The same properties manifest with similar magnitude around single cells.

Rho Kinase Inhibition by AT13148 Blocks Pancreatic Ductal Adenocarcinoma Invasion and Tumor Growth.

The high mortality of pancreatic cancer demands that new therapeutic avenues be developed. The orally available small-molecule inhibitor AT13148 potently inhibits ROCK1 and ROCK2 kinases that regulate the actomyosin cytoskeleton. We previously reported that ROCK kinase expression increases with human and mouse pancreatic cancer progression and that conditional ROCK activation accelerates mortality in a genetically modified LSL-KrasG12D; LSL-p53R172H; Pdx1-Cre; (KPC) mouse pancreatic cancer model. In this study, we show that treatment of KPC mouse and human TKCC5 patient-derived pancreatic tumor cells with AT13148, as well as the ROCK-selective inhibitors Y27632 and H1152, act comparably in blocking ROCK substrate phosphorylation. AT13148, Y27632, and H1152 induced morphologic changes and reduced cellular contractile force generation, motility on pliable discontinuous substrates, and three-dimensional collagen matrix invasion. AT13148 treatment reduced subcutaneous tumor growth and blocked invasion of healthy pancreatic tissue by KPC tumor cells in vivo without affecting proliferation, suggesting a role for local tissue invasion as a contributor to primary tumor growth. These results suggest that AT13148 has antitumor properties that may be beneficial in combination therapies or in the adjuvant setting to reduce pancreatic cancer cell invasion and slow primary tumor growth. AT13148 might also have the additional benefit of enabling tumor resection by maintaining separation between tumor and healthy tissue boundaries.Significance: Preclinical evaluation of a small-molecule ROCK inhibitor reveals significant effects on PDAC invasion and tumor growth, further validating ROCK kinases as viable therapeutic targets in pancreatic cancer. Cancer Res; 78(12); 3321-36. ©2018 AACR.