Proceedings of The Physiological Society

Europhysiology 2018 (London, UK) (2018) Proc Physiol Soc 41, PCA103

Poster Communications

Screening colorectal cancer cell lines for connexin-dependent intercellular diffusive coupling

S. Monterisi1,2, J. Michl1,2, W. F. Bodmer2, P. Swietach1

1. Department of Physiology, Anatomy & Genetics, University of Oxford, Oxford, United Kingdom. 2. Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom.

Background: According to a widely held canon, most types of cancer have little or no cell-to-cell diffusive coupling involving gap junctions (GJs) (1). Moreover, connexin (Cx) proteins, which assemble GJs, are suggested to be tumour suppressors, putatively because they restrict cell motility and limit the extent to which a cancer cell can regulate its cytoplasmic milieu independently of its neighbours (2). Recently, we have shown that some cancer cell lines, such as pancreatic Colo357, are diffusively coupled by Cx43 channels, which establishes a pathway for exchanging metabolites, including lactate and H+ ions (3). Here, we screened a panel of colorectal cancer (CRC) cell lines for evidence of coupling. Methods: A gene expression database from >100 CRC cell lines was searched for expression levels of connexin-coding genes (GJA1 for Cx43, GJB2 for Cx26, GJB3 for Cx31). Measurements of cell-to-cell coupling were performed on a selection of CRC cell lines that included various combinations of Cx isoform expression, ranging from high to low/absent. The protein products were tested by western blot and immunofluorescence. Diffusive coupling was quantified in calcein-loaded monolayers by fluorescence recovery after photobleaching (FRAP), which involves bleaching one cell in a monolayer and measuring the time constant of fluorescence recovery, which is then converted into a permeability constant. Results: The bioinformatics search revealed that the most substantial variation in expression was found for the gene coding for Cx43, a protein typically associated with neurons and cardiac muscle, and for Cx26, an isoform found in normal epithelia. CRC cells manifested different degrees of inter-cellular communication. Specifically, GJB2-expressing Cx26-positive CRC lines DLD1 and SNU1235 produced monolayers in which diffusive coupling was detected in the majority of the cells (permeability: 0.015±0.002, 0.012±0.002 µm/s, respectively). In contrast, GJA1-expressing Cx43-positive CRC lines CACO2, C10 and LOVO demonstrated heterogeneity, observed as distinct sub-populations with strong coupling or no coupling. The ratio of these sub-populations ranged from 1:1 in C10 and LOVO cells to 3:1 in CACO2 cells. Coupled cells demonstrated strong coupling, ranging from 0.010±0.001 µm/s in LOVO cells to 0.038±0.002 µm/s in CACO2 cells. Conclusions: Our results show that cell-to-cell connectivity varies between CRC cells and even within subpopulations of a CRC line. The extent of diffusive coupling between connected cells is sufficient to permit a substantial flux of small molecules, including metabolites. This could be particularly important in solid tumours where large diffusion gradients of metabolites, formed along gradients of oxygen, could drive net fluxes through coupled cells and hence influence tissue physiology.

Where applicable, experiments conform with Society ethical requirements