The goal of our work is to understand how cancer cells spread around the body. In order to address this problem we use a combination of in vivo tumour imaging, three-dimensional 'organotypic' models and conventional cell and molecular biology. Intravital imaging of tumours has revealed that only a subset of cells in primary tumours is motile and that this behaviour is not maintained at secondary sites. This could be because the activation of signalling pathways that promote cancer cell motility occurs locally and transiently. Furthermore, it has been suggested that motile cancer cells may become less differentiated or trans-differentiate.
To address these issues Silvia Giampieri and Sophie Pinner have developed strategies to monitor the activity of signalling pathways and the differentiation status of cells in both breast cancer and melanoma models. These two parallel projects have revealed an important role for transient activation of TGFbeta signalling. Together with Caroline Hill's laboratory (LRI Developmental Signalling Laboratory) we are currently investigating the molecular mechanism by which TGFbeta signalling affects cell motility and whether it affects the differentiation status of the motile sub-population of cells. Following the departure of Silvia and Sophie in the autumn, Cerys Manning is continuing some of this work.
Our work using 'organotypic' models has previously shown the critical role of matrix remodelling by stromal fibroblasts in generating paths through the extra-cellular matrix that cancer cells subsequently use to invade (Figure 1). Cedric Gaggioli has now developed a 96 well assay for matrix remodelling by carcinoma-associated fibroblasts and screened a moderate size chemical library for compounds that prevent matrix remodelling. This approach identified several compounds that could prevent stromal fibroblasts from promoting the invasion of carcinoma cells without compromising the viability of either cell type. Steven Hooper has investigated the molecular mechanism by which one class of compounds identified in this screen reduces matrix remodelling. He has found that HMG-CoA reductase inhibitors reduce the membrane targeting of Rab family small GTPases and has identified specific Rab proteins that are required in fibroblasts to promote matrix remodelling but seem dispensible in carcinoma cells. One of these is critical for the delivery of integrins to the plasma membrane and subsequent interactions between the fibroblast and matrix.
Figure 1. Matrix remodelling by a carcinoma associated fibroblasts: different components of the matrix are shown in white, green and blue with actin cytoskeleton of the fibroblast in red.
Cristina Hidalgo and Shahid Chaudhry have been searching for genes involved in the fibroblast-led collective invasion of carcinoma cells. Cristina has demonstrated a role for genes involved in control of cell polarity in the collective invasion of carcinoma cells. Many of these genes were first identified in developmental biology systems and Chris Madsen, who joined the lab in September, will further investigate the role of genes involved epithelial morphogenesis in Drosophila in the invasion of epithelial cancers (collaboration with Barry Thompson, LRI Epithelial Biology Laboratory). SiRNA screening of regulators of ubiquitination by Shahid Chaudhry has led to a focus on the role of NFkB signalling and inflammatory cytokines in fibroblast-led invasion. Shahid is currently investigating the molecular mechanism by which regulators of NFkB signalling affect squamous cell carcinoma invasion.
In addition to the collective invasion of cancer cells our group is also interested in the molecular mechanics of how single cancer cells move around. Sophie Pinner has identified a novel role for PDK1 in this process and has collaborated with Chris Marshall's group at the Institute for Cancer Research to study single cell motility in vivo. This work identified some of the control network that determines whether cancer cells move with an elongated morphology or a rounded morphology. Critically the morphology or 'mode' or invasion affects whether cells are sensitive to potential anti-invasion drugs such as the ROCK inhibitor Y27632. We are now hoping to model aspects of these changes in cell morphology and mode of migration in collaboration with Paul Bates (LRI Biomolecular Modelling Laboratory).
For a list of refereed research papers, see Publications (in navigation on left).
