Interfacial Organization and Forces Arising from Epithelial-Cancerous Monolayer Interactions.

The interfacial interactions between epithelia and cancer cells have profound relevance for tumor development and metastasis. Through monolayer confrontation of MCF10A (nontumorigenic human breast epithelial cells) and MDA-MB-231 (human epithelial breast cancer cells) cells, we investigate the epithelial-cancerous interfacial interactions at the tissue level. We show that the monolayer interaction leads to competitive interfacial morphodynamics and drives an intricate spatial organization of MCF10A cells into multicellular finger-like structures, which further branch into multiple subfinger-like structures. These hierarchical interfacial structures penetrate the cancer monolayer and can spontaneously segregate or even envelop cancer cell clusters, consistent with our theoretical prediction. By tracking the substrate displacements via embedded fluorescent nanobeads and implementing nanomechanical modeling that combines atomic force microscopy and finite element simulations, we computed mechanical force patterns, including traction forces and monolayer stresses, caused by the monolayer interaction. It is found that the heterogeneous mechanical forces accumulated in the monolayers are able to squeeze cancer cells, leading to three-dimensional interfacial bulges or cell extrusion, initiating the p53 apoptosis signaling pathways of cancer cells. We reveal that intercellular E-cadherin and P-cadherin of epithelial cells differentially regulate the interfacial organization including migration speed, directionality, spatial correlation, F-actin alignment, and subcellular protrusions of MCF10A cells; whereas E-cadherin governs interfacial geometry that is relevant to force localization and cancer cell extrusion, P-cadherin maintains interfacial integrity that enables long-range force transmission. Our findings suggest that the collaborative molecular and mechanical behaviors are crucial for preventing epithelial tissues from undergoing tumor invasion.

Collective Polarization of Cancer Cells at the Monolayer Boundary.

Cell polarization, a process depending on both intracellular and intercellular interactions, is crucial for collective cell migration that commonly emerges in embryonic development, tissue morphogenesis, wound healing and cancer metastasis. Although invasive cancer cells display weak cell-cell interactions, they can invade host tissues through a collective mode. Yet, how cancer cells without stable cell-cell junctions polarize collectively to migrate and invade is not fully understood. Here, using a wound-healing assay, we elucidate the polarization of carcinoma cells at the population level. We show that with loose intercellular connections, the highly polarized leader cells can induce the polarization of following cancer cells and subsequent transmission of polarity information by membrane protrusions, leading to gradient polarization at the monolayer boundary. Unlike the polarization of epithelial monolayer where Rac1/Cdc42 pathway functions primarily, our data show that collective polarization of carcinoma cells is predominantly controlled by Golgi apparatus, a disruption of which results in the destruction of collective polarization over a large scale. We reveal that the Golgi apparatus can sustain membrane protrusion formation, polarized secretion, intracellular trafficking, and F-actin polarization, which contribute to collective cancer cell polarization and its transmission between cells. These findings could advance our understanding of collective cancer invasion in tumors.

Mechanical adaptions of collective cells nearby free tissue boundaries.

Mechanical adaptions of cells, including stiffness variation, cytoskeleton remodeling, motion coordination, and shape changing, are essential for tissue morphogenesis, wound healing, and malignant progression. In this paper, we take confluent monolayers of Madin-Darby canine kidney (MDCK) and mouse myoblast (C2C12) cells as model systems to probe how cells collectively adapt their mechanical features in response to a free tissue boundary. We show that the free boundary not only can trigger unjamming transition but also induces cell fluidization nearby the boundary. The Young's moduli of cells near the boundary are found to be much lower than those of interior cells. We demonstrate that the stiffness of cells in monolayers with a free tissue boundary exhibits negative dependence on the projected cell area, in contrast to previous studies where cells were found to stiffen as cellular area increases in a confluent MDCK monolayer without boundary. In addition, the free tissue boundary may activate cell remodeling, rendering volume enlargement and cell-specified cytoskeleton organization. Our study emphasizes the important role of geometrical boundary in regulating biomechanical properties of cell aggregates.

DNAskew: statistical analysis of base compositional asymmetry and prediction of replication boundaries in the genome sequences.

Sueoka and Lobry declared respectively that, in the absence of bias between the two DNA strands for mutation and selection, the base composition within each strand should be A=T and C=G (this state is called Parity Rule type 2, PR2). However, the genome sequences of many bacteria, vertebrates and viruses showed asymmetries in base composition and gene direction. To determine the relationship of base composition skews with replication orientation, gene function, codon usage biases and phylogenetic evolution, in this paper a program called DNAskew was developed for the statistical analysis of strand asymmetry and codon composition bias in the DNA sequence. In addition, the program can also be used to predict the replication boundaries of genome sequences. The method builds on the fact that there are compositional asymmetries between the leading and the lagging strand for replication. DNAskew was written in Perl script language and implemented on the LINUX operating system. It works quickly with annotated or unannotated sequences in GBFF (GenBank flatfile) or fasta format. The source code is freely available for academic use at http://www.epizooty.com/pub/stat/DNAskew.