BACKGROUND/AIMS: Radixin has recently been shown to correlate with the metastasis of gastric cancer, but the pathogenesis is elusive. Adhesion proteins contribute to the regulation of metastasis, and thus this study sought to investigate the role of radixin in the migration, invasion and adhesion of gastric cancer cells, as well as its interaction with adhesion proteins in vitro. METHODS: Radixin stable knockdown human gastric carcinoma SGC-7901 cells were constructed. Alterations in the migration, invasion and adhesion ability were examined by matrigel-coated plate and transwell assays. The expression pattern of adhesion proteins, including E-cadherin, β-catenin and claudin-1, was determined by quantitative real-time PCR and western blot. Possible involvement of NF-κB/snail pathway was also evaluated. RESULTS: Stable knockdown of radixin significantly suppressed migration and invasion, but enhanced adhesion in SGC-7901 cells. The expression of E-cadherin was manifestly increased in radixin knockdown cells, whereas the expression of β-catenin and claudin-1 was unchanged. The nuclear exclusion of NF-κB followed by conspicuous reduction of snail expression was involved in the regulation of E-cadherin expression. CONCLUSIONS: Radixin knockdown suppresses the metastasis of SGC-7901 cells in vitro by up-regulation of E-cadherin. The NF-κB/snail pathway contributes to the regulation of E-cadherin in response to depletion of radixin.

Atlas-based automated anatomical labeling is a fundamental tool in medical image segmentation, as it defines regions of interest for subsequent analysis of structural and functional image data. The extensive investigation of multi-atlas warping and fusion techniques over the past 5 or more years has clearly demonstrated the advantages of consensus-based segmentation. However, the common approach is to use multiple atlases with a single registration method and parameter set, which is not necessarily optimal for every individual scan, anatomical region, and problem/data-type. Different registration criteria and parameter sets yield different solutions, each providing complementary information. Herein, we present a consensus labeling framework that generates a broad ensemble of labeled atlases in target image space via the use of several warping algorithms, regularization parameters, and atlases. The label fusion integrates two complementary sources of information: a local similarity ranking to select locally optimal atlases and a boundary modulation term to refine the segmentation consistently with the target image's intensity profile. The ensemble approach consistently outperforms segmentations using individual warping methods alone, achieving high accuracy on several benchmark datasets. The MUSE methodology has been used for processing thousands of scans from various datasets, producing robust and consistent results. MUSE is publicly available both as a downloadable software package, and as an application that can be run on the CBICA Image Processing Portal (https://ipp.cbica.upenn.edu), a web based platform for remote processing of medical images.