Intravital FRAP Imaging using an E-cadherin-GFP Mouse Reveals Disease- and Drug-Dependent Dynamic Regulation of Cell-Cell Junctions in Live Tissue

摘要

class="head no_bottom_margin" id="sec1title">IntroductionThe capacity of cancer cells to dissociate from primary tumors and invade requires the deregulation of interactions with adjacent cells and the surrounding tissue. A major challenge in biology is the real-time monitoring of protein dynamics involved in this process in their native context (). The ability to quantify the intricate spatiotemporal regulation of cell adhesion molecules, such as E-cadherin, using techniques including fluorescence recovery after photobleaching (FRAP) has rapidly enhanced our understanding of E-cadherin’s subcellular roles in regulating cell-cell integrity and dissociation in vitro (, , , ).FRAP is commonly used for monitoring molecular movement within cells. A small fluorescent region is bleached, and fluorescence recovery into the bleached region is measured over time (, , ). From this, multiple readouts can be derived, including, but not limited to the half-time of recovery, a measure of the rate at which fluorescent molecules move in or out of the region of interest, and the immobile fraction, an indication of how much of the molecule remains trapped and unable to move out of the analyzed region (for in-depth insights into FRAP analysis, see ). In the case of fluorescently labeled E-cadherin, the immobile fraction can indicate how much E-cadherin is trapped or engaged in cell-cell junctions and may provide a molecular readout of junction stability in real time (, ).FRAP has largely been used to probe molecular events within 2D cell culture using transfection-based approaches (, ), whereas its utility in vivo has been limited (). Recently, we and others have used FRAP in more complex and physiologically relevant environments ranging from application in Drosophila () to the use of E-cadherin-GFP FRAP in a mammalian system in vivo (href="#bib54" rid="bib54" class=" bibr popnode">Serrels et al., 2009), where E-cadherin-GFP-expressing squamous cell carcinoma cells were transplanted and grown subcutaneously in mice. Using this approach, we demonstrated that locally invading cells had a significantly lower immobile fraction of E-cadherin-GFP compared with non-invading cells, illustrating that, in cancer cells that retain E-cadherin expression, mobilization rather than loss of E-cadherin can play a role in reducing cell-cell adhesions, leading to more malleable, motile, and invasive tumor behavior (href="#bib20" rid="bib20" class=" bibr popnode">Friedl et al., 2012, href="#bib32" rid="bib32" class=" bibr popnode">Lamouille et al., 2014, href="#bib55" rid="bib55" class=" bibr popnode">Shamir et al., 2014). Although these approaches provide insights into the mobilization of E-cadherin in subcutaneous xenograft tumors, they lack the fidelity to recapitulate the complex and distinct microenvironment found in specific organs of interest (href="#bib13" rid="bib13" class=" bibr popnode">Conway et al., 2014). It is therefore essential to develop new tools for the quantification of molecular dynamics in distinct organs in which the disease of interest originates, allowing us to understand cell behavior at a subcellular and tissue- and disease-specific level.In this study, we generated a Cre-inducible E-cadherin-GFP mouse and exploited this model using intravital FRAP imaging to assess changes in E-cadherin-based cell-cell junction integrity during disease progression and in response to drug treatment. Here we mimic the disease etiology of pancreatic cancer in situ, from acquisition of an initiating Kras mutation that occurs in 95% of pancreatic cancers (href="#bib5" rid="bib5" class=" bibr popnode">Biankin et al., 2012, href="#bib26" rid="bib26" class=" bibr popnode">Hingorani et al., 2003) to subsequent loss- or gain-of-function mutations in the tumor suppressor p53, which occur in 50%–75% of pancreatic cancers (href="#bib5" rid="bib5" class=" bibr popnode">Biankin et al., 2012). Using FRAP analysis, we reveal that E-cadherin stability in normal pancreas or in non-invasive pancreatic tumors (Kras^G12D alone or Kras^G12D; p53^−/−) remains unaltered during disease progression. However, in mice with Kras^G12D and gain-of-function mutations in p53 (p53^R172H), E-cadherin is mobilized, facilitating the weakening of cell-cell contacts, correlating with the enhanced metastasis seen in this model (href="#bib27" rid="bib27" class=" bibr popnode">Hingorani et al., 2005, href="#bib40" rid="bib40" class=" bibr popnode">Morton et al., 2010b). Moreover, in line with recent work assessing Src kinase as an anti-invasive drug target (href="#bib2" rid="bib2" class=" bibr popnode">Avizienyte et al., 2002, href="#bib39" rid="bib39" class=" bibr popnode">Morton et al., 2010a, href="#bib45" rid="bib45" class=" bibr popnode">Nobis et al., 2013, href="#bib46" rid="bib46" class=" bibr popnode">Nobis et al., 2014), we demonstrate that the phase II drug dasatinib reverts E-cadherin mobilization in invasive pancreatic tumors and that this stabilization of junctions could partially explain its current anti-invasive role in this disease (T.J. Evans et al., 2012, ASCO, abstract). We therefore present the application of FRAP in the E-cadherin-GFP mouse for live, tissue-specific assessment of vital cell-cell adhesion changes in conjunction with its utility as a pre-clinical imaging tool for evaluating the efficacy of new therapeutics in the pancreas and other organs of interest in real time.