The KRAS-variant and its impact on normal breast epithelial cell biology

Abstract

MicroRNA (miRNA)-binding site variants in 3' untranslated regions (3'UTRs) are a novel class of germ-line, functional mutations, which are now recognized as powerful biomarkers of human cancer risk and biology. The first mutation discovered in this class is the KRAS-variant, a let-7-binding site mutation in the 3'UTR of the KRAS oncogene. The KRAS-variant predicts increased cancer risk for certain populations, is a predictive biomarker of cancer treatment response across cancer types, leads to conserved tumor biology and elevated AKT signaling in KRAS-variant patient tumors, and was recently found to predict elevated TGF-β and immunosuppression in cancer patients. Based on the functional biology of the KRAS-variant in cancer patients, here we chose to investigate altered normal cellular biology in the presence of the KRAS-variant, through interrogation of an isogenic normal breast epithelial cell line model with and without the KRAS-variant. We find that KRAS-variant normal breast epithelial cells exhibit a mesenchymal phenotype, which appears to be due to numerous molecular changes, including miRNA dysregulation and autocrine pathway alterations, including elevated TGF-β, resulting in ZEB and SNAIL upregulation. Our findings support the hypothesis that the KRAS-variant has a fundamental biological impact on normal cellular biology, that is conserved in these patients when they develop cancer.

Fig. 1

MicroRNA and phenotypic differences in KRAS-variant MT cells. a The KRAS-variant alters global microRNA expression. Representation of miRNA expression levels in MCF10A cells with (MCF10a^KRAS+/−; MT1 and MT2) versus without (MCF10a^KRAS−/−; WT) the KRAS-variant. b Relative mRNA expression of miR-200c in MCF10A cells with (MCF10a^KRAS+/−; MT1 and MT2) versus without (MCF10a^KRAS−/−; WT) the KRAS-variant at baseline as evaluated by qPCR. c Representative of phase contrast images (top; ×10 objective) and AlexaFluor 488-conjugated phalloidin staining of F-actin (bottom; ×40 objective) in MCF10A cells with (MT1 and MT2) versus without (WT) KRAS-variant grown as a monolayer. The size bars depict 100 and 50 µm length, respectively, in the images. Results of three independent experiments are shown (mean ± SEM). A Student’s t-test (two-tailed) was used to compare two groups and p < 0.05 was considered significant

Fig. 2

EMT in KRAS-variant epithelial cells. a The representative picture of immunofluorescence images (×40 objective) of MCF10a cells with (MCF10a^KRAS+/−; MT1 and MT2) versus without (MCF10a^KRAS−/−; WT) the KRAS-variant grown as a monolayer. Cells were stained to detect the expression of the epithelial marker, E-cadherin (red) and the mesenchymal marker, Fibronectin (green). Nuclei were stained with DAPI (blue). The expression of epithelial marker, E-cadherin (Red) is shown in MCF10A WT cells. Upregulation of the mesenchymal marker, Fibronectin (green) is visible in MCF10A MT1 and MT2. The size bars in yellow depict 50 µm length in the images. b Western blot analysis of E-cadherin and occludin (epithelial markers) and Fibronectin, vimentin and N-cadherin (mesenchymal markers) in MCF10A line with (MCF10A^KRAS+/−; MT1 and MT2) and without (MCF10A^KRAS−/−; WT) the KRAS-variant are shown. The expression of GAPDH was used as an internal control. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline using ImageJ software. c Western blot analysis of SNAIL, SLUG, ZEB1 TWIST1 and 2 (EMT transcription factors) in MCF10A line with (MCF10A^KRAS+/−; MT1 and MT2) and without (MCF10A^KRAS−/−; WT) KRAS-variant are shown. The expression of GAPDH was used as an internal control. Results of three independent experiments are shown (mean) and compared quantitatively using ImageJ software with GAPDH as baseline

Fig. 3

mir-200c and TGFβ lead to EMT through separate pathways in KRAS-variant cells. a Western blot analysis of EMT transcription factors ZEB1 and SNAIL in MCF10A line with (MT1 and MT2) and without (WT) the KRAS-variant after lentiviral transduction of control (pre-mir-vector) and pre-mir-200c are shown. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline using ImageJ software. b TGFβ1, TGFβ2, and TGFβ3 expression levels in MCF10A cells with (MCF10a^KRAS+/−; MT1 and MT2) and without (MCF10a^KRAS−/−; WT) the KRAS-variant are shown. Levels of TGFβ1 (Abcam), TGFβ2 (Abcam), TGFβ3, and GAPDH (Santa Cruz Biotechnology) were estimated by Western blot analysis. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline via ImageJ software. c Western blot analysis for detection of phosphorylation of Smad2, Smad3, ERK, AKT, and GAPDH in WT, MT1, and MT2 cells at baseline. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline via ImageJ software. d Impact of TGFβ pathway inhibition on EMT was determined by treating MCF10A WT, MT1 and MT2 cells with 10 µM of Galunisertib for 0, 24, 48, and 72 h. Levels of transcription factors ZEB1 and SNAIL and phosphorylation of AKT were measured by Western blotting. Results of three independent experiments are shown (mean) and compared quantitatively using ImageJ software with GAPDH as baseline. e Following serum starvation, cells were treated with 20 µM of LY294002 (PI3K inhibitor) for 1 h. Detection of phosphorylation of AKT and GSK3β and Snail, and GAPDH in WT, MT1 and MT2 was performed by Western blot analysis. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline

Fig. 4

Autocrine signaling in KRAS-variant cells. a The CXCR2 inhibitor SB225002 was used on MT cells at indicated doses for 1 h, and AKT phosphorylation was evaluated by western blot. Results of three independent experiments are shown (mean) and compared quantitatively with tAKT as baseline with the help of ImageJ software. b Conditioned medium (CM) from MCF10A cells with (MCF10^KRAS+/−; MT1 and MT2) the KRAS-variant was added to MCF10A WT cells that had been starved for 24 h in serum and growth factor-free medium. Whole cell lysates were prepared after the indicated times and analyzed by Western blotting. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline. c MCF10A cells (MCF10^KRAS+/−; MT1 and MT2) and (MCF10A ^KRAS−/−; WT) were grown together in chambered wells with serum starved growth factor-free medium so that MCF10A WT cells had continued access to growth factors exclusively secreted from MT cells for a time duration of 24 and 48 h. Whole cell lysates were prepared after the indicated times and analyzed by Western blotting. Results of three independent experiments are shown (mean) and compared quantitatively with GAPDH as baseline using ImageJ software

Fig. 5

A model of EMT in KRAS-variant MCF10A epithelial cells