Implication of Cytokeratin Expression during EMT

Abstract

Epithelial-mesenchymal transition (EMT) is the biological process of functional transformation of polarised epithelial cells into mesenchymal cells in which migratory properties are gained, and epithelial characteristics such as intercellular junctions, cell polarity and expression of epithelial markers are lost. It occurs in normal physiological development and tissue repair. However, in cancers, this developmental programme is hijacked by cancer cells to develop properties that are associated with increased levels of cell invasion, proliferation, migration, metastasis and apoptotic resistance. At the heart of the EMT process are a group of core regulators known as EMT transcription factors (EMT TFs), including SNAIL, SLUG, TWIST1, ZEB1, and ZEB2. The overexpression of EMT TFs has been used to recreate different EMTs in an in vitro cell culture model. The EMT TFs initiate the EMT programme and act on a wide array of downstream effectors including keratins, which are part of the intermediate filament system comprising the cellular cytoskeleton and can be divided into both type I (acidic) and type II ( neutral/basic) keratins. To maintain the cell's stability and provide the necessary support, the keratin cytoskeleton is constantly assembled and disassembled, ultimately impacting important cellular functions such as cell proliferation, migration and shape. Despite their importance, cytokeratins have been used in only a limited capacity in EMT studies, which have often mentioned only a single keratin, and more often than not using this keratin in a narrow scope as a marker of cell epithelial phenotype, with some studies not reporting on keratins at all. There does not appear to be any study that explores the impact of different keratins on the EMT programme. Therefore, this study aimed to overexpress EMT TFs in a single cell line expressing four different cytokeratins, and to assess the impact of different cytokeratin polypeptides on EMT through the assessment of cell proliferation, cell migration and changes in cell shape.

To select our cell line of interest, we used qPCR and western blot (WB) to screen for EMT TFs, cytokeratin 18 (K18), cytokeratin 19 (K19) and expression of the mesenchymal marker vimentin. The screening revealed that mesenchymal cell lines express EMT TFs at far higher levels than epithelial cells. Furthermore, it showed that epithelial and mesenchymal cells express both K18 and vimentin in distinctive patterns with respective negative and positive correlations to EMT transcription factors, which support their use as pan markers of differentiation between epithelial and mesenchymal phenotype within the EMT process. On the other hand, the expression of K19 in epithelial and mesenchymal cells is more complex and appears to be cell-type dependent. Therefore, it warrants further investigation to determine its role in the EMT process. The MCF-7 cell line was selected as the cell line for EMT TFs overexpression as it expressed EMT TFs and vimentin at lower levels of expression while at the same time expressing a wide spectrum of cytokeratins, including K8, K18, K19, K80 and its splice variant K80.1.

In addition to the wild-type MCF-7 cell line, two more variants were generated. In the first, K19 was knocked out using CRISPR-Cas9 technology to produce K19 deficient MCF-7 (K19KO). In the second, K80 was knocked down using shRNA (A3). K19 knockout downregulated K8, K80, and K80.1, while K80 knockdown did not impact K19. Thus, each of the wild-type MCF-7, K19KO, and A3 cell lines had its own distinctive expression of keratins. Using the MG132-specific ubiquitin-proteasome pathway inhibitor revealed that K80 is susceptible to ubiquitination control and that this control is dependent on the presence of K19. To assess keratins' impact on proliferation, MTT assay was used, while 2D migration assay was used to assess cell migration. K19 was found to induce proliferation and migration. The expression of K80, in the presence of K19, inhibited proliferation and promoted migration. To study the impact of K19, K80 and K80.1 on the cell shape, InCarta image analysis software was used. K19 expression was found to reduce nuclear and cell size, changing the shape of the cell to a rounder profile while also decreasing nuclear spacing. The expression of K80 had the opposite effect. Furthermore, the phenotypic data showed the expression of K80.1 has no effect on nuclear size or shape. This indicates that the extended tail domain of K80 might be specifically targeting the morphological features of the nucleus.

To induce EMT in our cell lines, EMT TFs were packaged and stably transduced into the MCF-7 cell line. The overexpression of wild-type SNAIL and SLUG failed to induce EMT, indicating that these TFs are under tight regulation in the MCF-7 cell line. The tagging of EMT TFs with an AcGFP tag at the N-terminus added stability to both ectopically expressed SNAIL and SLUG allowing their nuclear expression, contrary to C-terminus tagging. This could indicate that the AcGFP is having a masking effect on the SNAG domain (located on the N-terminus), which is shared by both these TFs, thus allowing the ectopic expression of SNAIL and SLUG. Only N-terminus tagged SLUG was able to induce a partial EMT, which was found to downregulate K80 translation but not transcription, indicating that the impact of SLUG-induced partial EMT on K80 is posttranslational. Future work should explore the expression of K8, K18, K19 and K80.1 in MCF-7 cells transduced with N-terminus tagged SLUG. Furthermore migration, proliferation, and phenotypic studies should be carried out and compared with K19KO and A3 cell lines, which are to be transduced with N-terminus tagged SLUG to determine the impact of having a different cytokeratin component on cells overexpressing EMT TFs.