Kinase GSK3β functions as a suppressor in colorectal carcinoma through the FTO‐mediated MZF1/c‐Myc axis

Abstract Colorectal carcinoma (CRC) poses heavy burden to human health and has an increasing incidence. Currently, the existing biomarkers for CRC bring about restrained clinical benefits. GSK3β is reported to be a novel therapeutic target for this disease but with undefined molecular mechanisms. Thus, we aimed to investigate the regulatory effect of GSK3β on CRC progression via FTO/MZF1/c‐Myc axis. Firstly, the expression patterns of GSK3β, FTO, MZF1 and c‐Myc were determined after sample collection. Lowly expressed GSK3β but highly expressed FTO, MZF1 and c‐Myc were found in CRC. After transfection of different overexpressed and interference plasmids, the underlying mechanisms concerning GSK3β in CRC cell functions were analysed. Additionally, the effect of GSK3β on FTO protein stability was assessed followed by detection of MZF1 m6A modification and MZF1‐FTO interaction. Mechanistically, GSK3β mediated ubiquitination of demethylase FTO to reduce FTO expression. Besides, GSK3β inhibited MZF1 expression by mediating FTO‐regulated m6A modification of MZF1 and then decreased the proto‐oncogene c‐Myc expression, thus hampering CRC cell proliferation. We also carried out in vivo experiment to verify the regulatory effect of GSK3β on CRC via FTO‐mediated MZF1/c‐Myc axis. It was found that GSK3β inhibited CRC growth in vivo which was reversed by overexpressing c‐Myc. Taken together, our findings indicate that GSK3β suppresses the progression of CRC through FTO‐regulated MZF1/c‐Myc axis, shedding light onto a new possible pathway by which GSK3β regulates CRC.

difficulty in its diagnosis. Apart from chemotherapy regimens, novel targeted therapies such as phosphoinositide 3-kinase (PI3K)-targeted therapy may be effective in the treatment of patients with CRC 5 and the glycogen synthase kinase 3 beta (GSK3β) mediates cell activities via the PI3K/AKT signalling pathway, 6 implying the regulatory role of GSK3β in CRC. However, the molecular mechanisms underlying the CRC initiation still remain undefined and current targeted agents bring restrained clinical benefit, so more innovative therapies targeting CRC are expected to be found for treating this malignant neoplasm.
Previous study has shown that the kinase GSK3β is an inhibitor gene for CRC occurrence. 7 Additionally, GSK3β could promote the ubiquitination of fat mass and obesity-associated protein (FTO) by regulating FTO phosphorylation, thereby elevating FTO degradation and suppressing FTO protein expression. 8 As a N6-methyladenosine (m6A) demethylase, FTO could inhibit the m6A methylation of mRNA, thereby affecting cell viability. 9 Moreover, FTO was found to be overexpressed in the cells of patients with gastric cancer, 10 suggesting its oncogenic role. Specifically, FTO is reported to remove m6A mRNA methylation of myeloid zinc finger 1 (MZF1), thus increasing the expression of MZF1. 11 MZF1 is a carcinogenic transcription factor that regulates the invasion of multiple solid cancers including CRC, cervical cancer and hepatocellular carcinoma, and it has also been implicated in PI3K pathway. 12 Hence, we speculated that inhibition of MZF1 can serve as an efficient therapy to prevent CRC progression, and the kinase GSK3β may inhibit the expression of FTO and MZF1, thus functioning as a suppressor of CRC. To prove the hypothesis and explore the molecular mechanisms, we have performed a series of experiments.

| Clinical sample collection
Colorectal carcinoma tissues and adjacent normal tissues were excised from 57 patients with CRC diagnosed in Linyi People' s Hospital from August 2018 to August 2019 and stored in liquid nitrogen for subsequent research. None of these patients received radiation therapy or pre-operative chemotherapy. Sigma-Aldrich Chemical Company) supplemented with 10% foetal bovine serum, 100 ug/mL penicillin and 50 ug/mL streptomycin in a constant temperature incubator with 5% CO 2 at 37°C. After that, the cells in adherent growth were detached with 0.25% trypsin (T4799, Sigma), followed by collection of the cells in logarithmic growth for subsequent experiments. The overexpressed plasmid pCMV6-AC-GFP and the interference plasmid pGPU6/Neo were purchased from Fenghuishengwu Co., Ltd. and GenePharma Co. Ltd., respectively. The cells were

| Reverse transcription quantitative polymerase chain reaction (RT-qPCR)
Total RNA was isolated from tissues and cells using TRIzol (Invitrogen), and the concentration and purity of which were measured using a NanoDrop 2000 micro ultraviolet spectrophotometer (1011U; NanoDrop Technologies Inc.). The RNA was reversely transcribed into cDNA according to the instructions of PrimeScript reagent Kit (RR047A; Takara Holdings Inc.), and primers for MZF1 were designed and synthesized by Takara (Table 1). ABI7500 qPCR instrument (7500; Applied Biosystems) was employed for real-time fluorescent qPCR detection. β-actin was used as an internal reference, and the fold changes were calculated by means of relative quantification (2 -△△Ct method).

| Western blot analysis
Total protein in tissues or cells was extracted with radio immunoprecipitation assay (RIPA) lysis buffer containing phenylmethanesulfonyl fluoride (P0013C; Beyotime Institute of Biotechnology), with the concentration detected by BCA kit.
After sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), the protein was electrotransferred onto a polyvinylidene fluoride membrane and blocked with 5% skim milk powder at room temperature for 1 hour. Then, the membrane was probed with diluted primary antibodies to GSK3β (1:5000, Next day, the membrane was re-probed with horseradish peroxidase (HRP)-labelled secondary antibody of goat anti-rabbit antibody to immunoglobulin G (IgG) for 1 hour and visualized by enhanced chemiluminescence (ECL) kit (BB-3501, Ameshame; Chiltem). Finally, Quantity One v4.6.2 software was used to quantify the grey levels of each band in the Western blot image.
GAPDH served as an internal reference.

| Co-immunoprecipitation (Co-IP) assay
Cells were lysed on ice using IP lysis buffer (Wuxi Biogoodland Biotechnology Co., Ltd.) supplemented with protease inhibitor. Then, 1 mg protein was taken from per sample and immunoprecipitated with monoclonal antibody to FTO for incubation overnight at 4℃.
Next morning, 20 μL Protein A + G beads were added for 2 hours incubation. After that, the precipitates were eluted with IP lysis, and the sample was centrifuged at 2500 rpm for 5 minutes at 4℃ to dis-

| Protein stability determination
Cyclohexanamide (CHX) treatment: To determine the FTO protein stability, the protein was extracted after centrifugation of CRC cells lysed with RIPA lysis buffer (P0013B; Beyotime) at 12 000 r/ min and incubated with the protein synthesis inhibitor cyclohexanamide (CHX, 100 mg/mL) for 0 hour, 1 hour or 2 hours before analysis.
MG132 treatment: Cells were treated by 20 mM MG132 for 12 hours and then collected. Then, the cells were lysed in RIPA buffer containing 0.1% SDS and immunoblotted. ImageJ was used to quantify FTO expression normalized to GAPDH. The protein was extracted at 0 hour, 1 hour and 2 hours, and the expression of FTO was measured by Western blot analysis. The proteasome inhibitor MG132 (20 mM) and the CHX (100 mg/mL) were purchased from Sigma.
Next, isolated and purified mRNA and magnetic bead-antibody complex were added into the IP buffer containing RNase inhibitor and protease inhibitor for incubation at 4°C overnight. The RNA was eluted, and MZF1 was analysed by RT-qPCR after purification using phenol-chloroform extraction.

| Photoactivatable-ribonucleoside-enhanced cross-linking and immunoprecipitation (PAR-CLIP)
Colorectal carcinoma cells were cultured with 200 μM 4-nitropyridine (Sigma) for 14 hours and cross-linked at 365 nm with 0.4 J/cm 2 . Immunoprecipitation was carried out after lysis with FTO antibody overnight at 4°C. Then, the precipitated RNA was labelled with (γ-32-P)-adenosine triphosphate (ATP) and visualized through autoradiography. For RT-qPCR analysis, the precipitate was first detached with proteinase K to remove protein, and then, RNA extracted by TRIzol was used to perform RT-qPCR to measure the expression of MZF1.

| Cell counting kit-8 (CCK-8) method
The CRC cell proliferation was counted using the CCK-8 (CK04, in the dark, the flow cytometer was employed to detect apoptosis.

| Xenograft tumour in nude mice
Twenty-four specific pathogen free female BALB/c nude mice (age:

| Statistical analysis
The SPSS 21.0 version (IBM Corp.) was used for statistical analysis.
The measurement data were presented by mean ± standard deviation.
The data of CRC tissues and adjacent normal tissues were analysed by paired t test, and the data of the other two groups were compared by unpaired Student's t test. The data among groups were analysed using one-way analysis of variance (ANOVA) and Tukey's post hoc test, and the cell experimental data at different time points were compared using two-way ANOVA. Tumour data at different time points were analysed by repeated measures ANOVA followed by Bonferroni's post hoc test. Statistical significance was assumed at P < .05.

| GSK3β inhibited proliferation of CRC cells by down-regulating FTO expression
We collected clinical samples of CRC, cultured CRC cells (SW480,

SW620 and HCT-8 cells) and HIEC cells and performed Western blot
and RT-qPCR experiments to detect the expression of GSK3β and FTO in CRC tissues and cells. The results showed ( Figure 1A,B) that the expression of GSK3β was significantly reduced in CRC tissues and cells, while the expression of FTO was significantly increased. Additionally, SW620 cells were used later due to the lowest GSK3β expression and the highest FTO expression among three CRC cells. In order to study whether GSK3β has an effect on the biological characteristics of CRC cells through mediating the expression of FTO, the SW620 cells were transfected with oe-GSK3β or oe-FTO. Western blot analysis showed ( Figure 1C) overexpression of GSK3β significantly decreased the expression of FTO, but GSK3β was not affected by promoted FTO. After GSK3β was overexpressed, up-regulating FTO could recover the expression of FTO but had no significant effect on GSK3β.
CCK-8 assay was performed to detect the proliferation of SW620 and HCT-8 cells in each group, and the results indicated ( Figure 1D;

| GSK3β inhibited MZF1 expression by mediating FTO-mediated m6A modification of MZF1 in CRC cells
Previous study has shown that FTO promotes MZF1 expression by removing m6A mRNA modification of MZF1. 11 First, the expression of MZF1 in CRC tissues and cells was assessed by Western blot analysis, and the results showed ( Figure 3A,B) that MZF1 expression was significantly higher in CRC tissues and cells. Next, the Western blot analysis showed ( Figure 3C, Figure S1B) that the F I G U R E 2 GSK3β inhibits the expression of m6A demethylase FTO in CRC cells. A, Western blot analysis of GSK3β expression after oe-GSK3β treatment. B, Co-IP detection of FTO ubiquitination in SW620 cells. C, Western blot analysis of FTO protein expression after treatment of oe-GSK3β combined with proteasome inhibitor MG132. D, Western blot analysis of FTO protein stability after treatment of oe-GSK3β combined with CHX. GAPDH was used as the internal reference. The data were expressed as mean ± standard deviation. The data of CRC tissues and adjacent normal tissues in normal distribution were analysed using paired t test, the data among groups were compared using oneway ANOVA, and the data among multiple groups at different time were analysed by two-way ANOVA. * P < .05 The data were expressed as mean ± standard deviation. The data of CRC tissues and adjacent normal tissues in normal distribution were analysed using paired t test, the data among groups were compared using one-way ANOVA, and the data among multiple groups at different time were analysed by two-way ANOVA. * P < .05

| FTO reduced c-Myc expression by downregulating MZF1
Existing literature has shown that MZF1 could promote the expres-

| FTO activated MZF1/c-Myc axis to promote CRC cell proliferation
Furthermore, the functions of FTO in SW620 cell biological processes were analysed with the involvement of MZF1/c-Myc axis.
The results of CCK-8 for detecting the proliferation of SW620 and HCT-8 cells revealed ( Figure 5A; Figure S1E) that low expression of FTO noticeably reduced the proliferative potential of SW620 cells, while overexpression of MZF1 increased the cell proliferation. The GAPDH was employed as the internal reference. The data were expressed as mean ± standard deviation. The data of CRC tissues and adjacent normal tissues in normal distribution were analysed using paired t test, the data among groups were compared using one-way ANOVA, and the data among multiple groups at different time were analysed by two-way ANOVA.

| GSK3β inhibited the growth of CRC xenografts by suppressing the expression of c-Myc
In order to further study the effect of GSK3β on the growth of CRC Collectively, GSK3β could reduce the expression of MZF1 through restraining FTO expression and thus lead to a decrease in the c-MYC expression, thus inhibiting the proliferative and migratory phenotypes of CRC cells while promoting their apoptosis.

| CON CLUS IONS
In conclusion, this study finds a novel GSK3β/FTO/MZF1/c-MYC axis which is involved in the occurrence and development of CRC and, clinically, proposes a therapeutic regimen based on using GSK3β as the inhibitor for treating CRC patients, which may hold great potential to relieve CRC-related pathogenic symptoms. In the future, we will study in more detail the role of GSK3β on CRC through FTO-regulated MZF1/c-MYC axis by adding different cell lines used in experiments, etc, and seek to find more effective therapies for treating this disease as the next action.

ACK N OWLED G EM ENTS
We would like to thank all participants that enrolled in the present study.

CO N FLI C T O F I NTE R E S T
All the authors declare that they have no competing interests.