Author response: KSR1- and ERK-dependent translational regulation of the epithelial-to-mesenchymal transition

Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract The epithelial-to-mesenchymal transition (EMT) is considered a transcriptional process that induces a switch in cells from a polarized state to a migratory phenotype. Here, we show that KSR1 and ERK promote EMT-like phenotype through the preferential translation of Epithelial-Stromal Interaction 1 (EPSTI1), which is required to induce the switch from E- to N-cadherin and coordinate migratory and invasive behavior. EPSTI1 is overexpressed in human colorectal cancer (CRC) cells. Disruption of KSR1 or EPSTI1 significantly impairs cell migration and invasion in vitro, and reverses EMT-like phenotype, in part, by decreasing the expression of N-cadherin and the transcriptional repressors of E-cadherin expression, ZEB1 and Slug. In CRC cells lacking KSR1, ectopic EPSTI1 expression restored the E- to N-cadherin switch, migration, invasion, and anchorage-independent growth. KSR1-dependent induction of EMT-like phenotype via selective translation of mRNAs reveals its underappreciated role in remodeling the translational landscape of CRC cells to promote their migratory and invasive behavior. eLife digest The majority of cancer deaths result from tumor cells spreading to other parts of the body via a process known as metastasis. 90% of all cancers originate in epithelial cells that line the inner and outer surface of organs in our bodies. Epithelial cells, however, are typically stationary and must undergo various chemical and physical changes to transform in to migratory cells that can invade other tissues. This transformation process alters the amount of protein cells use to interact with one another. For example, epithelial cells from the colon produce less of a protein called E-cadherin as they transition into migrating cancer cells and make another protein called N-cadherin instead. A protein called KSR1 is a key component of a signaling pathway that is responsible for generating the proteins colon cancer cells need to survive. But it is unknown which proteins KSR1 helps synthesize and whether it plays a role in the metastasis of colon cancer cells. To investigate this, Rao et al. studied the proteins generated by cancerous colon cells cultured in the laboratory, in the presence and absence of KSR1. The experiment showed that KSR1 increases the levels of a protein called EPSTI1, which colon cancer cells need to transform into migratory cells. Depleting KSR1 caused cancer cells to generate less EPSTI1 and to share more features with healthy cells, such as higher levels of E-cadherin on their surface and reduced mobility. Adding EPSTI1 to the cancer cells that lacked KSR1 restored the traits associated with metastasis, such as high levels of N-cadherin, and allowed the cells to move more easily. These findings suggest that KSR1 and EPSTI1 could be new drug targets for reducing, or potentially reversing, the invasive behavior of colon cancer cells. However, further investigation is needed to reveal how EPSTI1 is generated and how this protein helps colon cancer cells move and invade other tissues. Introduction Molecular scaffolds affect the intensity and duration of signaling pathways by coordinating a discrete set of effectors at defined subcellular locations to regulate multiple cell fates (Morrison and Davis, 2003; Pawson and Scott, 1997). Kinase Suppressor of Ras 1 (KSR1) serves as a scaffold for Raf, MEK, and ERK enabling the efficient transmission of signals within the mitogen activated protein kinase (MAPK) cascade (Kortum and Lewis, 2004; Nguyen et al., 2002). Although KSR1 is dispensable for normal development, it is necessary for oncogenic Ras-induced tumorigenesis including colorectal cancer cells (Kortum and Lewis, 2004; Nguyen et al., 2002; Fisher et al., 2011; Fisher et al., 2015; Morrison et al., 2009; Rao et al., 2020), suggesting that KSR1 may modulate aberrant signals that redirect the function of effectors typically involved in normal cellular homeostasis. Activating Ras mutations are present in over 40 % of colorectal cancers (CRC), and associated with advanced disease and decreased overall survival (Haigis, 2017; Serebriiskii et al., 2019). Activated Ras, a critical driver of both tumor growth and survival, is an alluring therapeutic target, yet targeting the majority of oncogenic Ras alleles is still a work in progress. Raf/MEK/ERK signaling can phenocopy Ras signaling essential for CRC growth and survival (Schmitz et al., 2007; Brandt et al., 2019). Therefore, understanding the effectors that transmit signals emanating from oncogenic Ras is a valuable step in detecting and targeting the pathways critical to tumor cell function and their adaptation to therapy. Oncogene-driven signaling pathways promote mRNA translation that enables expression of a subset of mRNAs to promote growth, invasion, and metastasis (Chu et al., 2016; Avdulov et al., 2004; Truitt Morgan et al., 2015; Pelletier et al., 2015). Tumor cells have an increased dependence on cap-dependent translation, unlike their normal complements (Truitt Morgan et al., 2015; Truitt and Ruggero, 2016). Eukaryotic Translation Initiation Factor 4E (eIF4E) is a rate-limiting factor for oncogenic transformation, with reductions of as little as 40 % being sufficient to block tumorigenesis (Truitt Morgan et al., 2015). eIF4E function is regulated by association of 4E-binding proteins (4EBPs). Importantly, disruption of KSR1 or ERK inhibition leads to dephosphorylation and activation of 4EBP1, indicating that the function of KSR1 as an ERK scaffold is key to the aberrant regulation of mRNA translation (McCall et al., 2016). This tumor-specific, KSR1-dependent regulation of mRNA translation of a subset of genes was predicted to selectively promote survival of CRC cells but not normal colon epithelia (McCall et al., 2016; Neilsen et al., 2019). Almost all CRC originates from epithelial cells lining the colon or rectum of the gastrointestinal tract, but in order to invade to the surrounding tissue, cancer cells lose cell adhesiveness to acquire motility and become invasive, characterized by the epithelial-to-mesenchymal transition (EMT), which is central to tumor pathogenesis (Ye and Weinberg, 2015; Nieto, 2013; Thiery et al., 2009; Dongre and Weinberg, 2019). EMT involves a complex cellular process during which epithelial cells lose polarity, cell-cell contacts and acquire mesenchymal characteristics. While EMT is crucial for cell plasticity during embryonic development, trans differentiation and wound healing, when aberrantly activated EMT has deleterious effects, which facilitate motility and invasion of cancer cells (Nieto, 2013; Thiery et al., 2009; Dongre and Weinberg, 2019; Nieto et al., 2016). EMT has been shown to be controlled by transcription-dependent mechanisms, especially through repression of genes that are hallmarks of epithelial phenotype such as E-cadherin. Loss of E-cadherin at the membrane has been associated with carcinoma progression and EMT (Thiery et al., 2009; Thiery, 2002; Oda et al., 1994; Frixen et al., 1991). E-cadherin function is transcriptionally repressed through the action of EMT transcription factors (TFs), including Snail-family proteins (Snail1, Slug), zinc finger E-box binding homeobox 1 and 2 (ZEB1 and ZEB2), and twist-related protein (Twist) (Nieto et al., 2016; Jolly et al., 2017). Transcriptional control of E-cadherin is unlikely to be sole determinant of EMT, invasion and metastasis. Inappropriate induction of non-epithelial cadherins, such as N-cadherin by epithelial cells are known to play a fundamental role during initiation of metastasis (Nieman et al., 1999; Liu et al., 2017; Suyama et al., 2002; Rosivatz et al., 2004; Okubo et al., 2017; Sadot et al., 1998; Loh et al., 2019). N-cadherin disassembles adherent junction complexes, disrupting the intercellular cohesion and reorienting the migration of cells, away from the direction of cell-cell contact (Nieman et al., 1999; Scarpa et al., 2015). Upregulation of N-cadherin expression promotes motility and invasion (Nieman et al., 1999; Liu et al., 2017; Suyama et al., 2002; Hulit et al., 2007). Thus, central to the process of EMT is the coordinated loss of E-cadherin expression and the upregulation of N-cadherin gene expression, termed cadherin switching (Loh et al., 2019; Wheelock et al., 2008; Tomita et al., 2000; Maeda et al., 2005; Araki et al., 2011). Previous studies have demonstrated transcriptional regulation of EMT through oncogenic Ras or its downstream effector signaling pathways via the activation of EMT-TFs (Shin et al., 2010; Shin et al., 2019; Andreolas et al., 2008; Liu et al., 2014; Wong et al., 2013; Wang et al., 2010; Lemieux et al., 2009). Oncogenic Ras itself activates EMT-TF Slug to induce EMT in skin and colon cancer cells (Wong et al., 2013; Wang et al., 2010). Enhanced activity of ERK2 but not ERK1, has been linked with Ras-dependent regulation of EMT (Shin et al., 2010; Shin et al., 2019). Several studies have also described an alternative program wherein cells lose their epithelial phenotype, via post-transcriptional modifications rather than transcriptional repression involving translational regulation or protein internalization (Jechlinger et al., 2003; Aiello et al., 2018; Waerner et al., 2006). Expression profiling of polysome-bound mRNA to assess translational efficiency identified over 30 genes that were translationally regulated upon Ras and TGFβ inducing EMT (Jechlinger et al., 2003; Waerner et al., 2006). Functional characterization of the resultant proteins should reveal preferentially translated mRNAs essential to invasion and metastasis. EPSTI1 was identified as a stromal fibroblast-induced gene upon co-cultures of breast cancer cells with stomal fibroblasts (Nielsen et al., 2002). EPSTI1 is expressed at low levels in normal breast and colon tissue but aberrantly expressed in breast tumor tissue (Nielsen et al., 2002). EPSTI1 promotes cell invasion and malignant growth of primary breast tumor cells (Li et al., 2014; de Neergaard et al., 2010). We performed polysome profiling in CRC cells and found that KSR1- and ERK induces of EPSTI1 mRNA translation. EPSTI1 is both necessary and sufficient for coordinating the upregulation of N-cadherin with the downregulation of E-cadherin to stimulate cell motility and invasion in colon cancer cells. These data demonstrate that ERK-regulated regulation of mRNA translation is an essential contributor to EMT-like phenotype and reveal a novel effector of the cadherin switch whose characterization should yield novel insights into the mechanisms controlling the migratory and invasive behavior of cells. Results Genome-wide polysome profiling reveals translational regulation of EPSTI1 by KSR1 ERK signaling regulates global and selective mRNA translation through RSK1/2-dependent modification of cap-dependent translation (McCall et al., 2016; Roux et al., 2007). Phosphorylation of cap binding protein 4E-BP1 releases eIF4E to promote translation and the abundance of eIF4E is a rate-limiting factor for oncogenic Ras- and Myc-driven transformation (Truitt Morgan et al., 2015). We showed previously that KSR1 maximizes ERK activation in the setting of oncogenic Ras (Kortum et al., 2006), which is required for increased Myc translation via dephosphorylation of 4E-BP1, supporting CRC cell growth (McCall et al., 2016). These observations imply that the ERK scaffold function of KSR1 alters the translational landscape in CRC cells to support their survival. To determine the effect of KSR1 on translatomes in colon cancer cells, we performed genome-wide polysome profiling (King and Gerber, 2016). We stably expressed short hairpin RNA (shRNA) constructs targeting KSR1 (KSR1 RNAi) or a non-targeting control in two K-Ras mutant CRC cell lines, HCT116 and HCT15 (Figure 1D, top panels). We isolated and quantified both total mRNA and efficiently translated mRNAs (associated with ≥3 ribosomes) using RNA sequencing (Figure 1A, Figure 1—figure supplement 1). We used Anota2seq (Oertlin et al., 2019) to calculate translation efficiency (TE) by comparing the differences in efficiently translated mRNAs to the total transcript of each mRNA and observed that a significant number of mRNAs ([selDeltaTP ≥ log (1.2) and selDeltaPT ≥ log (1.2)] and p-value < 0.05) showed either reduced TE or upregulated TE upon KSR1 disruption (Figure 1B–C, Supplementary file 1, Source data 1) in both HCT116 and HCT15 cells. Gene Set Enrichment Analysis (GSEA) (Subramanian et al., 2005) of significantly enriched genes in HCT116 and HCT15 (Figure 1B, Figure 1—figure supplement 2A-B), identified 11 mRNAs in the gene set titled "Hallmark EMT signature", "Jechlinger EMT Up", and "Gotzmann EMT up" , that had significantly decreased translation upon KSR1 disruption (Supplementary file 2). Among the genes with decreased translation, EPSTI1 was one of the highly significant mRNAs. We sought to determine the functional relevance of KSR1-dependent induction of EPSTI1 to phenotypic plasticity in colon cancer cells. Figure 1 with 2 supplements see all Download asset Open asset EPSTI1 translation is regulated by KSR1. (A) Representative polysome profiles from control and KSR1 knockdown (KSR1 RNAi) HCT116 and HCT15 cells. Sucrose gradient fractions 3–5 denote the low-molecular-weight complexes (monosomes) and the fractions 6–8 are the high-molecular-weight complexes (polysomes). (B) Scatter plot of polysome-associated mRNA to total mRNA log2 fold-changes upon KSR1 knockdown in HCT116 and HCT15 with RNA-seq. The statistically significant genes in the absence of KSR1 are classified into four groups with a fold change (|log2FC|) > 1.2 and p-value < 0.05. The number of mRNAs with a change in TE (orange and red) are indicated (n = 3 for each condition). TE, translational efficiency. (C) Heatmap of TE changes for the top 40 RNAs control and KSR1 knockdown (KSR1 RNAi) HCT116 and HCT15 cells (n = 3 for each condition). (D) Western blot analysis of KSR1 and EPSTI1 following KSR1 knockdown in HCT116 and HCT15 cells. (E) RT-qPCR analysis of EPSTI1 mRNA from total RNA and polysomal RNA (fractions number 6–8) in control and KSR1 knockdown HCT116 and HCT15 cells, the TE was calculated as the ratio of polysomal mRNA to the total mRNA (n = 3; *, p < 0.05). (F) RT-qPCR analysis of EPSTI1 mRNA levels isolated from sucrose gradient fractions of the control and KSR1 knockdown HCT116 and HCT15 cells. Fractions 3–5 (low MW) and 6–8 (high MW) are plotted for the control and KSR1 knockdown state with values corresponding to the percentage of total mRNA across these fractions n = 3. Experiments shown in (A - F) are representative of three independent experiments. To confirm that EPSTI1 translation is KSR1-dependent, we observed that, EPSTI1 protein expression was decreased with the knockdown of KSR1 in HCT116 and HCT15 cells (Figure 1D), while the total mRNA transcript was unchanged upon KSR1 disruption (Figure 1E, left panel). EPSTI1 TE was markedly decreased upon KSR1 depletion (Figure 1E, right). RT-qPCR analysis of sucrose-gradient fractions of monosome mRNA and polysome RNA distribution confirmed that EPSTI1 mRNA shifted from actively translating high-molecular-weight (MW) polysome fractions to low-MW fractions in KSR1 knockdown cells (Figure 1F). In contrast, HPRT1 mRNA was insensitive to KSR1 knockdown in HCT116 and HCT15 cells, and qPCR analysis of HPRT1 mRNA isolated from sucrose gradient fractions of control and KSR1 knockdown cells showed no significant shift between the low-MW and the high-MW fractions (Figure 1—figure supplement 2C). To determine if KSR1 promotes EPSTI1 degradation, we first assessed EPSTI1 turnover in HCT116 cells following treatment with a protein-synthesis inhibitor, cycloheximide (CHX) and observed that EPSTI1 has a 6 hr half-life (Figure 1—figure supplement 2D). We analyzed EPSTI1 turnover using a combination of proteasome inhibitor, MG132 and CHX in control and CRISPR-targeted KSR1 HCT116 cells (Figure 1—figure supplement 2E). EPSTI1 turnover was not sensitive to MG132 treatment in HCT116 cells lacking KSR1 expression. Therefore, in HCT116 cells, KSR1 does not mediate ubiquitin proteosome system (UPS)-mediated degradation of EPSTI1. These data support our conclusion that EPSTI1 translation is induced by KSR1. KSR1/ERK signaling regulates EPSTI1 expression in colon cancer cells To confirm our observations in KSR1 knockdown cells, we tested the effect of CRISPR/Cas9-mediated targeting of KSR1 on EPSTI1 in CRC cell lines. EPSTI1 protein expression was decreased upon KSR1 depletion in HCT116 and HCT15 cells and EPSTI1 expression was restored in knockout cells upon expression of a KSR1 transgene (+ KSR1) (Figure 2A). Similar to inhibition of KSR1, treatment with ERK inhibitor SCH772984 (Morris et al., 2013) suppressed EPSTI1 protein expression in both CRC cell line HCT116 and tumorigenic patient derived colon organoid engineered with deletion of APC, p53, SMAD4, and K-RasG12D mutation (PDO-11 AKPS) (Figure 2B; Drost et al., 2015). To determine if EPSTI1 expression is also dependent on mTOR signaling, we tested the effect of mTOR inhibition on EPSTI1expression. Though mTOR inhibitor, AZD8055 (Chresta et al., 2010) robustly inhibited phosphorylation of mTOR substrate p70 S6 kinase, its ability to decrease EPSTI1 expression in HCT116 cells was weak relative to treatment with the ERK inhibitor (Figure 2C). These observations suggest the ERK affects EPSTI1 expression via mechanisms distinct from mTOR. While the total protein was reduced upon ERK inhibition in HCT116, the EPSTI1 transcript levels were not altered significantly by SCH772984 treatment (Figure 2D). Figure 2 Download asset Open asset KSR1 or ERK inhibition suppresses EPSTI1 protein expression in cell lines and organoids. (A) Cell lysates prepared from control, KSR1 CRISPR-targeted (KSR1 CRISPR) and CRISPR-targeted HCT116 and HCT15 cells expressing KSR1 (KSR1 CRISPR+ KSR1) analyzed for EPSTI1 protein expression by western blotting. (B) Western blot of the indicated proteins in HCT116 (left) and AKPS quadruple mutant organoids (right) treated with DMSO or 1 µM of SCH772984 for 48 hr. (C) EPSTI1 protein expression was analyzed by western blot in HCT116 cells treated with DMSO, 1 µM of SCH772984, or 1 µM of AZD8055 for 48 hr. (D) RT-qPCR analysis of EPSTI1 mRNA from total RNA in HCT116 cells treated with either DMSO or ERK1/2 selective inhibitor, SCH772984 (n = 3; ns, non-significant). (E) Representative polysome profiles from HCT116 cells treated DMSO or 1 µM of ERK1/2 selective inhibitor, SCH772984. (F) RT-qPCR analysis of EPSTI1 and HPRT1 mRNA levels from LMW (fractions 3–5) and HMW (fractions 6–8) of the DMSO control or SCH772984-treated HCT116 cells (n = 3; *, p < 0.05; ***, p < 0.001). All values displayed as mean ± S.D. Experiments shown in (A - F) are representative of three independent experiments. We performed polysome profiling in HCT116 cells, either treated with DMSO or ERK inhibitor, SCH772984 and we isolated mRNA from low-MW monosome (fractions 3–5) and high-MW polysome (fractions 6–8) fractions (Figure 2E). RT-qPCR demonstrated that EPSTI1 mRNA shifted from high-MW fractions to the low-MW fractions upon ERK inhibition (Figure 2F). The distribution of mRNA for HPRT1 within the same profile was not altered by SCH772984 treatment (Figure 2F). These data indicate that KSR1-dependent ERK signaling is a critical regulator of EPSTI1 mRNA translation in colon cells and organoids. EPSTI1 is required for anchorage-independent growth in colon cancer cells KSR1 disruption inhibits HCT116 cell anchorage-independent growth in vitro and tumor formation in vivo (Fisher et al., 2015). Similarly, disruption of KSR1 by CRISPR/Cas9-mediated targeting decreased HCT116 and HCT15 cell viability under anchorage-independent conditions on simulated by poly-(HEMA) coating (Figure 3A). KSR1 transgene expression restored cell viability in HCT116 and HCT15 cells lacking KSR1 (KSR1 CRISPR+ KSR1) (Figure 3A). We showed previously that KSR1 expression is upregulated in colon cancer cell lines when to the human colon epithelial cells (Fisher et al., 2015). We observed that EPSTI1 protein is aberrantly expressed in colon cancer cell lines HCT116 and while its expression is in (Figure EPSTI1 protein expression is also markedly higher in AKPS organoids than normal colon organoids (Figure Figure 3 Download asset Open asset EPSTI1 is overexpressed in cancer cell lines and organoids and promotes anchorage-independent growth. (A) cell viability was analyzed in HCT116 and HCT15 cells on was using following (KSR1 CRISPR) and KSR1 (KSR1 CRISPR+ KSR1) in the CRISPR-targeted cells. The data are shown as relative mean ± n = were analyzed for by Western blot the expression of KSR1 in control, KSR1 knockout and cells expressing a KSR1 transgene (+ (B) Western blot analysis of EPSTI1 protein expression was assessed in HCT116, normal human colon and AKPS colon organoids. (C) of HCT116 and HCT15 cells using following knockdown of EPSTI1 that were on to anchorage-independent Cell viability was and 1, and 3 (n = The data are shown as mean ± were analyzed for by Western blot the knockdown of EPSTI1 in HCT116 and at 3. (D) of the in HCT116 and cells following knockdown using non-targeting control or EPSTI1 on Representative of for each The data are as the number of present 2 mean ± n = were analyzed for using Western blot the knockdown of EPSTI1 in HCT116 and cells. p < To determine the regulation of EPSTI1 in human colon tumor we performed knockdown of EPSTI1 in HCT116 and HCT15 cells. EPSTI1 disruption suppressed viability on poly-(HEMA) by 40 % in HCT15 cells, and over in HCT116 cells (Figure EPSTI1 knockdown reduced formation in by % in HCT116 cells and % in cells (Figure These observations show that KSR1-dependent translation of is required for anchorage-independent growth of colon tumor cell lines. KSR1 or EPSTI1 disruption cell in CRC cells the role of EPSTI1 in EMT-like (Nielsen et al., 2002; et al., we sought to the role of EPSTI1 in colon cancer cells. of control and EPSTI1 knockdown in HCT116 cell motility in a wound was analyzed by the relative wound et al., over (Figure was used to calculate relative wound that the percentage of cell the wound relative to the of the wound at a The of cell migration using this changes in cell to was also assessed in control, CRISPR-targeted (KSR1 and HCT116 cells expressing KSR1 (KSR1 CRISPR+ KSR1) (Figure lacking either EPSTI1 or KSR1 were % less to control cells. of KSR1 expression in CRISPR-targeted HCT116 cells restored motility to the control cells (Figure Figure Download asset Open asset KSR1 or EPSTI1 promote migration and invasion in CRC cells. (A) CRISPR-targeted (KSR1 CRISPR) and CRISPR-targeted HCT116 cells expressing KSR1 (KSR1 CRISPR+ KSR1) and control or EPSTI1 knockdown HCT116 cells were in a wound The the of wound calculated by shown as mean ± n = p < were analyzed for using with for multiple (B) KSR1 knockout (KSR1 and EPSTI1 knockdown were to migration through for hr using % as The number of cells were Data are the mean ± (n = *, p < p < ***, p < Representative of cells hr invasion through EPSTI1 knockdown HCT116 and cells were to invasion EPSTI1 suppresses cell invasion through by % in HCT116 and by % in (Figure top and KSR1 is required for EPSTI1 translation, we the functional of KSR1 in cell KSR1 depletion suppressed invasion by % in HCT116 and by % cells (Figure top left and these suggest the KSR1-dependent EPSTI1 signaling to cell migration and invasion in CRC cells. KSR1 or EPSTI1 disruption cadherin switching in CRC cells To the by which KSR1 or EPSTI1 promote motility and invasion in CRC cells, we their to the expression of critical of EMT that modulate cell E- and and to the non-targeting control, KSR1 disruption in HCT116, HCT15 and cells had levels of with a decrease in EMT-TF Slug (Figure Expression of and was not in HCT116 cells (Figure supplement knockdown of EPSTI1 with either of two we observed a decrease in the expression of N-cadherin, ZEB1 and Slug. with the decrease in E-cadherin levels were (Figure While was no significant change in the Slug and ZEB1 mRNA upon EPSTI1 knockdown (Figure supplement EPSTI1 disruption decreased N-cadherin mRNA expression over % in HCT116 and cells (Figure EPSTI1 we control HCT116 cells and HCT116 cells N-cadherin to invasion through EPSTI1 knockdown suppressed cell The expression of N-cadherin in cells lacking EPSTI1 was sufficient to to HCT116 cells (Figure supplement This is with observations that upregulation of N-cadherin expression motility in multiple cancer cell lines (Nieman et al., 1999; Hulit et al., 2007; et al., These indicate that the switch of E-cadherin to N-cadherin expression promotes the progression of migratory and invasive behavior by EPSTI1 signaling in CRC cells. Figure with 1 supplement see all Download asset Open asset KSR1 and EPSTI1 promote cadherin (A) Western blot analysis of the cell lysates prepared from control, and two of CRISPR-targeted HCT116, and HCT15 cells (KSR1 CRISPR) for the and EPSTI1. (B) Western blot of and N-cadherin in HCT116 and cells hr following EPSTI1 (C) RT-qPCR analysis of EPSTI1 mRNA and N-cadherin following knockdown of EPSTI1 for hr in HCT116 and cells. n = ***, p < p < Western shown in (A) and (B) and qPCR shown in (C) are representative of at three independent experiments. EPSTI1 is necessary and sufficient for EMT-like phenotype in CRC cells To determine the to which KSR1- and EPSTI1 translation is critical to colon tumor cell growth and invasion, we expressed a in knockout HCT116, and HCT15 cells. CRISPR/Cas9-mediated deletion of KSR1 EPSTI1 expression, Slug and N-cadherin expression and E-cadherin expression (Figure E-cadherin was in control CRC cells but at the cell membrane in KSR1 knockout cells (Figure expression of EPSTI1 in cells lacking KSR1 restored the cadherin switch, by decreasing the expression of E-cadherin (Figure and and N-cadherin levels to control cells (Figure of E-cadherin and of N-cadherin expression by the EPSTI1 transgene the ability of KSR1 knockout cells to in (Figure and invade through expression of EPSTI1 in these cells, increased the number of cells by over (Figure To determine the effect of EPSTI1 on cell we analyzed the cell growth in HCT116 and cells (Figure supplement 1). 3 EPSTI1 knockdown had no effect on cell to control HCT116 and cells. While EPSTI1 expression in KSR1 knockout cells had no significant effect on cell for hr in HCT116 an