Aberrant expression of embryonic mesendoderm factor MESP1 promotes tumorigenesis

Abstract

Background: Mesoderm Posterior 1 (MESP1) belongs to the family of basic helix-loop-helix transcription factors. It is a master regulator of mesendoderm development, leading to formation of organs such as heart and lung. However, its role in adult pathophysiology remains unknown. Here, we report for the first time a previously-unknown association of MESP1 with non-small cell lung cancer (NSCLC).

Methods: MESP1 mRNA and protein levels were measured in NSCLC-derived cells by qPCR and immunoblotting respectively. Colony formation assay, colorimetric cell proliferation assay and soft agar colony formation assays were used to assess the effects of MESP1 knockdown and overexpression in vitro. RNA-sequencing and chromatin immunoprecipitation (ChIP)-qPCR were used to determine direct target genes of MESP1. Subcutaneous injection of MESP1-depleted NSCLC cells in immuno-compromised mice was done to study the effects of MESP1 mediated tumor formation in vivo.

Findings: We found that MESP1 expression correlates with poor prognosis in NSCLC patients, and is critical for proliferation and survival of NSCLC-derived cells, thus implicating MESP1 as a lung cancer oncogene. Ectopic MESP1 expression cooperates with loss of tumor suppressor ARF to transform murine fibroblasts. Xenografts from MESP1-depleted cells showed decreased tumor growth in vivo. Global transcriptome analysis revealed a MESP1 DNA-binding-dependent gene signature associated with various hallmarks of cancer, suggesting that transcription activity of MESP1 is most likely responsible for its oncogenic abilities.

Interpretation: Our study demonstrates MESP1 as a previously-unknown lineage-survival oncogene in NSCLC which may serve as a potential prognostic marker and therapeutic target for lung cancer in the future.

Keywords: ARF; Lineage-survival oncogene; Lung cancer; Mesp1.

Fig. 1

MESP1 expression is associated with non-small cell lung cancer. (a) MESP1 transcript expression in various normal (grey colored) versus cancerous patients (orange colored) and (b) MESP1 transcript expression in normal versus lung adenocarcinoma patients (various stages) (top); MESP1 expression in normal and lung squamous cell carcinoma patients (various stages) (bottom), from TCGA and GTEx datasets. (c) qRT-PCR analysis of MESP1 mRNA in RNA samples from normal lung, lung squamous cell carcinoma and lung adenocarcinoma patients. (d) qRT-PCR analysis of MESP1 mRNA and immunoblots (low and high exposure) of MESP1 to check for endogenous level of hMESP1 in Beas-2B, H358, A549, H1944, H1299 and H460 cell lines. β−ACTIN was used as the housekeeping gene for normalization in qRT-PCR analysis and also as a loading control for western blot. The upper band in MESP1 immunoblot represents a likely modification. (e) Representative images at 5X and 40X magnification (right) of MESP1-immunohistochemical (IHC) staining from lung tissue arrays, including tissues from normal lung, normal adjacent tumor, lung adenocarcinoma, and lung squamous cell carcinoma. IHC scores are determined by staining intensity and percentage of stained cells; (left) quantification according to cytoplasmic IHC expression of MESP1 in tumor specimens including normal lung and normal adjacent tumor tissues. In all panels, data are represented as mean ± SD. * p < 0.05; ** p < 0.01; *** p < 0.001 (by two-tailed Student's t-test). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).

Fig. 2

MESP1 regulates growth and cell proliferation of NSCLC cells. (a) qRT-PCR and immunoblot of hMESP1 in shControl, shMESP1#5 and shMESP1#6 in A549 and H358 cells. β-Actin serves as the loading control. (b) Cell proliferation analysis by MTS assay of shControl, shMESP1#5 and shMESP1#6 in A549 and H358 cell lines. (c) Colony formation assay of shControl, shMESP1#5 and shMESP1#6 in A549 and H358 cells and quantification of colonies per well ± SD (bottom). (d) Schematic representation of the DNA-binding mutant of MESP1, with amino-acid residues responsible for DNA-binding shown by vertical lines (left), and immunoblot (right) of A549 cells transiently expressing MESP1 (FLAG-tagged) for control, wt-MESP1 and EK-mutant-MESP1. (e ) Cell proliferation analysis by MTS. (f) Colony formation assay of control, wt-MESP1 and EK-mutant-MESP1 A549 cells (left) with quantification of colonies per field (right). In all panels, data are represented as mean ± SD. * p < 0.05; ** p < 0.01; *** p < 0.001 (by two-tailed Student's t-test).

Fig. 3

MESP1 causes oncogenic transformation of ARF null MEFs. (a) Schematic representation of the DNA-binding mutant of Mesp1, with amino-acid residues responsible for DNA-binding shown by vertical lines (left); qRT-PCR and immunoblot (right) of Mesp1 (V5-tag) for control, wt-Mesp1 and EK-mutant-Mesp1 ARF -/- MEFs. (b and c) Cell proliferation analysis by MTS (b) and colony formation assay (c) of control, wt-Mesp1 and EK-mutant-Mesp1 ARF -/- MEFs. (d) Anchorage-independent growth of control, wt-Mesp1 and EK-mutant-Mesp1 ARF -/- MEFs as determined using a soft agar colony formation assay. Representative images of colonies in soft agar (left) and quantification of the colonies per field (right). Scale bar: 2 mm (low magnification) and 50 μm (higher magnification). In all panels, data are represented as mean ± SD. * p < 0.05; ** p < 0.01; *** p < 0.001 (by two-tailed Student's t-test).

Fig. 4

MESP1 dependent genes encompass various hallmarks of cancer. (a) Heat map of differentially regulated genes (DEGs) in control, wt-Mesp1 and EK-mutant-Mesp1 p19^ARF -/- MEFs. (b) Heat map of differentially regulated genes in shControl and shMESP1#6 expressing A549 cells. (c and d) Table (c) and Venn diagram (d) showing an overall overlap between DEGs upon Mesp1-overexpression (MESP1-OE) in ARF -/- MEFs and MESP1-knockdown (MESP1-KD) in A549 cells and bar graphs for mSigDB-GSEA hallmark analysis of respective gene lists (Bottom).

Fig. 5

Identification of 24 MESP1 target genes that predict poor prognosis. (a) Venn diagram (left) and bar graph for mSigDB-GSEA hallmark analysis (right) of overlapping genes across genes upregulated by Mesp1 overexpression in ARF -/- MEFs and genes-downregulated by MESP1-knockdown in A549 cells; and heat map of overlapping genes in corresponding systems (bottom). (b, c and d) qRT-PCR analysis of corresponding genes upon MESP1 knockdown in A549 (b) and H358 (c) cells and upon Mesp1-overexpression (d) in ARF -/- MEFs. (e) ChIP assay of endogenous MESP1 for target genes, presented as fold enrichment for MESP1 antibody over IgG antibody control in H358 cells. (f) Kaplan‐Meier Curve for overall survival of LUSC patients segregated by low or high expression of the 24 gene signature together with MESP1. In all panels, data are represented as mean ± SD. * p < 0.05; ** p < 0.01; *** p < 0.001 (by two-tailed Student's t-test).

Fig. 6

MESP1 knockdown cells have reduced tumor formation. (a, b and c) Tumor weight (a), tumor volume (b) and tumor images (c) after 38 days post subcutaneous injection of shControl, shMESP1#5 and shMESP1#6 stably transduced A549 cells in NSG mice. (d) H&E staining of tumors obtained from NSG mice injected with shControl, shMESP1#5 and shMESP1#6 stably transduced A549 cells. (e and f) Tumor weight (e) and H&E staining (f) of tumors obtained from NSG mice after 15 days post subcutaneous injection of shControl, shMESP1#5 and shMESP1#6 stably transduced H358 cells. (g) Schematic representation of MESP1’s function as a lineage-survival oncogene in lung cancer. In all panels, data are represented as mean ± SD. * p < 0.05; ** p < 0.01; *** p < 0.001 (by two-tailed Student's t-test).