Mindin serves as a tumour suppressor gene during colon cancer progression through MAPK/ERK signalling pathway in mice

Abstract

Mindin is important in broad spectrum of immune responses. On the other hand, we previously reported that mindin attenuated human colon cancer development by blocking angiogenesis through Egr-1-mediated regulation. However, the mice original mindin directly suppressed the syngenic colorectal cancer (CRC) growth in our recent study and we aimed to further define the role of mindin during CRC development in mice. We established the mouse syngeneic CRC CMT93 and CT26 WT cell lines with stable mindin knock-down or overexpression. These cells were also subcutaneously injected into C57BL/6 and BALB/c mice as well as established a colitis-associated colorectal cancer (CAC) mouse model treated with lentiviral-based overexpression and knocked-down of mindin. Furthermore, we generated mindin knockout mice using a CRISPR-Cas9 system with CAC model. Our data showed that overexpression of mindin suppressed cell proliferation in both of CMT93 and CT26 WT colon cancer cell lines, while the silencing of mindin promoted in vitro cell proliferation via the ERK and c-Fos pathways and cell cycle control. Moreover, the overexpression of mindin significantly suppressed in vivo tumour growth in both the subcutaneous transplantation and the AOM/DSS-induced CAC models. Consistently, the silencing of mindin reversed these in vivo observations. Expectedly, the tumour growth was promoted in the CAC model on mindin-deficient mice. Thus, mindin plays a direct tumour suppressive function during colon cancer progression and suggesting that mindin might be exploited as a therapeutic target for CRC.

Keywords: MAPK/ERK; colorectal cancer; mindin.

FIGURE 1

Mindin suppresses colon cancer cell proliferation in vitro. A, Analysis of CT26 WT (left side) and CMT93 (right side) cell proliferation in the mindin‐overexpressing cells and control cells by CCK‐8 assay (*P < 0.05). B, Analysis of cell proliferation in the mindin knock‐down cells and control cells by CCK‐8 assay (*P < 0.05). C, Analysis of cell proliferation in the mindin‐overexpressing cells and control cells by BrdU assay (*P < 0.05). D, Analysis of cell proliferation in the mindin knock‐down cells and control cells by BrdU assay (*P < 0.05)

FIGURE 2

Subcutaneous implantation tumour growth of mindin‐overexpressing CMT93 and CT26 WT cells. C57BL/6 and BALB/c mice were subcutaneously injected with stable mindin‐overexpressing CMT93 or CT26 WT cells, or PCMV4 control cells. Tumour size was measured every 3 d for 24 d. A, Images of isolated tumours from the four groups of study mice (n = 5). B, In vivo tumour growth resulting from the mindin‐overexpressing CMT93 or CT26 WT cell (n = 5, *P < 0.05). C, Western blot analysis confirming mindin protein overexpression in the tumour tissues of the four study groups at the end of the study period. Tubulin was used as a protein loading control (n = 5). Upper panel indicates CMT93, and lower panel indicates CT26 WT among A‐C

FIGURE 3

Subcutaneous implantation tumour growth of mindin knock‐down CMT93 and CT26 WT cells. C57BL/6 and BALB/c mice were subcutaneously injected with stable mindin knock‐down CMT93 or CT26 WT cells, or PU6 control cells. Tumour size was measured every 3 d for 24 d. A, Images of isolated tumours from the four groups of study mice (n = 5). B, In vivo tumour growth resulting from the mindin knock‐down CMT93 or CT26 WT cell (n = 5, *P < 0.05). C, Western blot analysis confirming mindin protein deficiency in the tumour tissues of the four study groups at the end of the study period. Tubulin was used as a protein loading control (n = 5). Upper panel indicates CMT93, and lower panel indicates CT26 WT among A‐C

FIGURE 4

Lentivirus‐mediated colitis‐associated cancer model. A, Experimental protocol used to induce CAC and the administration of lentiviral vectors. B, Images of isolated colon tissue from the mindin‐overexpression groups and control mice at the end of the study period (upper panel, n = 8), and tumour size and number (lower panel, n = 8, *P < 0.05, **P < 0.01). C, Photograph of isolated colon tissue from the mindin knock‐down groups and control mice (upper panel, n = 4), and tumour size and number (lower panel, n = 4, *P < 0.05, **P < 0.01). D and E, Confocal microscopy and anti‐mindin immunohistochemistry analysis of frozen and paraffin‐embedded colon sections showing the GFP reporter for lentiviral vector expression and mindin protein (as shown by the red arrows), and H&E staining of serial sections of mouse CAC tissues

FIGURE 5

Mindin‐mediated signalling pathway analysis. A, Western blot analysis using antibodies against ERK1/2, phospho‐ERK1/2, p65‐NF‐κB and phospho‐p65‐NF‐κB and protein lysates from mindin, PCMV4, shMindin and PU6 cells. GAPDH was used as a loading control. B, Western blot analysis of the phosphorylation level of ERK1/2 in mindin‐overexpressing (upper panel) or knock‐down (lower panel) tumour tissues from tumour subcutaneous implantation model mice. Tubulin was used as a protein loading control (n = 5). C and D, Western blot analysis of c‐Fos, FosB, c‐Jun, FRA1, CDK4, CDK6, CyclinD1, CyclinD3, P15 and P27 expression in the stable cell lines and their controls

FIGURE 6

U0126 inhibition of ERK1/2 phosphorylation, cell proliferation and colitis‐associated cancer model of mindin‐knockout mice. A, Western blot analysis of U0126‐treated cells using antibodies against ERK1/2 and phospho‐ERK1/2. GAPDH was used as a loading control. B, Analysis of U0126‐treated cell proliferation in the mindin‐overexpressing (left panel) or knock‐down (right panel) and control cells by BrdU assay (*P < 0.05). C, Tumour number (left panel) and size (right panel) of isolated colon tissue from the mindin‐knockout groups and control mice at the end of the study (n = 8, *P < 0.05, ***P < 0.01). D, Representative images of isolated colon tissue from the mindin‐knockout groups and control mice at the end of the study (left panel). Western blot analysis of the phosphorylation level of ERK in mindin‐knockout or control tumour tissues from CRC model mice (right panel). Tubulin was used as a loading control

FIGURE 7

A, Sequencing chromatograms show the nucleotide mutation of mindin−/− mice using a CRISPR‐Cas system. B, Western blot analysis using antibody against mindin on mice colon tissues. Actin was used as a loading control. C, The tumour images of 21 d after subcutaneous injection of CMT93 colorectal cancer cells in mindin‐knockout and the control mice. Tumour size was measured and quantitatively analysed (n = 5, *P < 0.05). D, Mindin expression was measured by ELISA in the first day and the end point of the model in WT mice serum with or without CRC procedure. E, The cells were isolated from tumour tissues of the AOM/DSS‐induced CRC mice. The procedure of flow cytometry analysis as follows: gated the single cells first, separated cells with the LIVE/DEAD dye and gated the CD45⁺ cells, then right panels showed the FITC‐CD11b and PE‐F4/80‐stained cells