Cell fate factor DACH1 represses YB-1-mediated oncogenic transcription and translation

Abstract

The epithelial-mesenchymal transition (EMT) enhances cellular invasiveness and confers tumor cells with cancer stem cell-like characteristics, through transcriptional and translational mechanisms. The mechanisms maintaining transcriptional and translational repression of EMT and cellular invasion are poorly understood. Herein, the cell fate determination factor Dachshund (DACH1), suppressed EMT via repression of cytoplasmic translational induction of Snail by inactivating the Y box-binding protein (YB-1). In the nucleus, DACH1 antagonized YB-1-mediated oncogenic transcriptional modules governing cell invasion. DACH1 blocked YB-1-induced mammary tumor growth and EMT in mice. In basal-like breast cancer, the reduced expression of DACH1 and increased YB-1 correlated with poor metastasis-free survival. The loss of DACH1 suppression of both cytoplasmic translational and nuclear transcriptional events governing EMT and tumor invasion may contribute to poor prognosis in basal-like forms of breast cancer, a relatively aggressive disease subtype.

Figure 1. DACH1 binding requires the YB-1 Cold Shock Domain (CSD)

A) Schematic representation of DACH1, YB-1 and YB-1 mutant expression vectors. B) IP-Western blot was conducted of HEK293T cells transfected with expression vectors encoding either N-terminal 3x FLAG-tagged YB-1 or HA-tagged DACH1 expression vectors. In (B) the IP was conducted with an antibody to the FLAG-tag of YB-1 and in (C) the IP was directed to the HA tag of DACH1. DACH1 binding to YB-1 was derived from N=3 separate experiments. (D) DACH1-YB-1 binding characterized in pulldown using 6-His-tagged DACH1 or DACH1 DBD region and GST-YB-1. (E–G) Schematic representation of YB-1 and DACH1 mutant expression vectors. (F–H) IP-Western blot of Myc-tagged YB-1 and associated DACH1 expression vectors transfected into HEK293T cells. Antibodies were directed for IP to the 3x FLAG tag of DACH1 and Western blot to the Myc epitope of YB-1. Antibodies are directed to the proteins as indicated. Data are representative of N=3 separate experiments. The DACH1 isoforms are indicated as DACH1a, b and c.

Figure 2. DACH1 inhibits YB-1 induced EMT

A) MDA-MB-231 cells stably transduced with a Ponasterone A-inducible FLAG-tagged DACH1 expression vector were sequentially transduced with shRNA to YB-1. Pon A treatment (2 μM) was used to induce DACH1. Western blot was conducted with the antibodies as indicated. (B–D) Immunohistochemistry for markers of mesenchymal cells include, (B) ZEB1, (C) Snail.

Figure 3. DAHC1 inhibits YB-1 nuclear transcriptional signaling pathways

(A) MDA-MB-231 cells expressing a DACH1 expression vector and/or YB-1 shRNA were analyzed by (A, E) RT-PCR or (B–D) genome wide microarray analysis. The relative changes in gene expression are shown. The -fold change or as a color scale of relative change in gene expression. The genes within DAVID signaling pathways for (B) “tumor cell movement” (C) “tumor cell proliferation” and (D) “EMT pathways” are shown for N=3 separate experiments. Note in (B) and (C) KLF-4 is a component of both DAVID pathways. Note in (D), no significant alterations in gene expression of the EMT pathway were detected. (E) Cellular proliferation and (F) tumor growth rates of MDA-MB-231 cells, expressing Ponasterone A inducible DACH1 with or without shRNA to YB-1, implanted into athymic nude mice. (G) Relative mRNA expression of KLF4 in MDA-MB-231 cells with etopic expression of DACH1 or shYB-1. (H) Immunohistochemistry of MDA-MB-231 cells shows relative KLF-4 abundance in ponasterone-inducible DACH1 stable MDA-MB-231 cells transduced with shRNA for YB-1. Cells were treated with Ponasterone A (2 μM) for 48 hours and shYB-1 for 48 hours. Antibodies are to proteins as indicated in the figure.

Figure 4. DACH1 antagonizes YB-1 induced migration and stem cell function

MDA-MB-231 cells, encoding Ponasterone A on a inducible DACH1, with or without YB-1 shRNA, were analyzed for migration, invasion and stem cell function using (A) transwell migration assays, (B) matrigel invasion assays and (C) 3D collagen invasion assays with the number of migrating or invading cells throughout shown as mean ±SEM for N=3 separate experiments. In (D), the same cell lines were used to determine mammosphere number and (E) mammosphere size with data shown throughout as mean ±SEM for N=5 separate experiments.

Figure 5. DACH1 promotes nuclear translocation of YB-1, inhibiting YB-1 dependent translation

Confocal microscopic images of SUM159 (CA) and MDA-MB-231 cells (B) with ectopic expression of DACH1 using antibodies as indicated. (C) Western blot analysis of cytoplasmic and nuclear protein as indicated. (D) MCF10A-RAS cell transfected with YB-1, DACH1 and SNAIL-5′UTR Luc reporter were starved and treated with rapamycin (uM). Luciferase activity was measured after 24 hours. The data are mean ±SEM.

Figure 6. YB-1 and DACH1 expression is inversely correlated in human Luminal A and Basal Genetic subtypes

The mRNA abundance for YB-1 and DACH1 in (A) basal breast cancer cell lines and (B) human breast cancer cell line. C) Immunohistochemistry for DACH1 and YB-1 in normal breast and basal breast cancer (B) mRNA for YB-1 and DACH1 in A) basal like breast cancer cell lines and B) human breast cancer cell lines. D) A combined breast cancer microarray data set assigning the five breast cancer genetic expression subtypes is shown as a heat map. The ESR1, epidermal-growth factor receptor (ERBB2) and progesterone receptor (PGR), is shown together with YB-1 and DACH1 expression. E–I) Normalized DACH1 expression plotted versus normalized YB-1 expression level reveals the inverse relationship between DACH1 and YB-1 expression in luminal A and basal genetic subtype (red circle).