The Wilms' tumor 1 (WT1) gene is well known as a chameleon gene. It plays a role as a tumor suppressor in Wilms' tumor but also acts as an oncogene in other cancers. Previously, our group reported that a canonical AUG starting site for the WT1 protein (augWT1) acts as a tumor suppressor, whereas a CUG starting site for the WT1 protein (cugWT1) functions as an oncogene. In this study, we report an oncogenic role of cugWT1 in the AOM/DSS-induced colon cancer mouse model and in a urethane-induced lung cancer model in mice lacking cugWT1. Development of chemically-induced tumors was significantly depressed in cugWT1-deficient mice. Moreover, glycogen synthase kinase 3β promoted phosphorylation of cugWT1 at S64, resulting in ubiquitination and degradation of the cugWT1 associated with the F-box−/− WD repeat-containing protein 8. Overall, our findings suggest that inhibition of cugWT1 expression provides a potential candidate target for therapy.

Significance:

These findings demonstrate that CUG-translated WT1 plays an oncogenic role in vivo, and GSK3β-mediated phosphorylation of cugWT1 induces its ubiquitination and degradation in concert with FBXW8.

The Wilms' tumor 1 (WT1) gene was originally discovered as a tumor suppressor gene responsible for Wilms' tumor, which is a neoplasm mostly affecting children (1). In contrast, as an oncogene, WT1 maintains the survival and metastasis of cancer cells by regulating various important genes, such as cyclin D1 (2) and Bcl2 (3). The WT1 gene is overexpressed in a wide variety of solid tumors, including lung (4), colon (5), esophageal (6), and breast (7, 8). It is a marker of poor prognosis and plays an important role in tumorigenesis (9). These reports suggest that WT1 acts a chameleon gene both as a tumor suppressor and as an oncogene (10).

WT1 protein isoforms arise from two translation initiation sites. The sites include the canonical AUG starting site for the WT1 protein (augWT1) and a CUG translation starting site 5′ to the traditional AUG initiation site. The CUG site produces a WT1 protein with an additional 68 amino acids at the N terminus (cugWT1; refs. 11, 12). However, previous WT1 functional studies did not distinguish between augWT1 and cugWT1 and thus functional differences between the variants are still unclear. Recently our group reported that augWT1 (52–56 kD) plays a role as a tumor suppressor, whereas, cugWT1 (62–68 kD) is expressed at high levels in various cancer cells or tissues and acts as an oncogene (13).

On the basis of our reported findings showing that cugWT1, but not augWT1, is the major form in various cancer cells, we decided to study posttranslational modifications affecting the turnover of cugWT1. Notably, only augWT1 has been reported to be subjected to posttranslational modifications of phosphorylation, ubiquitination, and sumoylation (14–16). We recently discovered that cugWT1 protein stability was increased by Akt's phosphorylation of WT1 at S68 (We reported phosphorylation of WT1 at S62, NP_077742.2. However NCBI recent updated a new sequence number for cugWT1, NP_077742.3.), which is located within the additional 73 amino acids of cugWT1. The phosphorylation of cugWT1 (S68) causes resistance to ubiquitination and degradation (13), but the mechanism is not fully understood.

Glycogen synthase kinase 3β (GSK3β) is a proline-directed serine/threonine kinase. Many transcription factors susceptible to phosphorylation-directed degradation are GSK3β substrates that cooperate with F-box−/− WD repeat-containing proteins (FBXW; refs. 17, 18). Moreover, various reports indicate that GSK3β can act as a potential therapeutic target in many types of cancers (19–21). On the basis of these reports, we speculated that a novel phosphorylation site on cugWT1 could be phosphorylated by GSK3β, in association with FBXWs proteins, to promote turnover.

In this study, we report an oncogenic role of cugWT1 in the AOM/DSS-induced colon cancer and urethane-induced lung cancer models. In these models, mice lacking cugWT1 were compared with wild-type (WT) mice. Both colorectal and pulmonary cancers are one of the most common fatal malignancies worldwide (22, 23). We determined the mechanism of cugWT1 degradation in carcinogenesis.

Reagents and antibodies

Cell culture media were obtained from Corning. Penicillin/streptomycin and FBS were from Gemini Bio-Products. Antibodies to detect WT1 6F-H2 (05-753/MAB4234-C) were purchased from Millipore and Flag (F3165) was from Sigma-Aldrich. Antibodies to detect GAPDH (sc-32233), Bcl-2 (sc-7382), cyclin D1 (sc-246), and EGFR (sc-373746) were obtained from Santa Cruz Biotechnology, Inc. Antibodies against p-GSK3β Ser9 (#9336), GSK3β (#9315), WT1 D8I7F (#83535), and PCNA (#13110) were from Cell Signaling Technology. V5 (MA5-15253) and HA-HRP (12013819001) were from Invitrogen and Roche, respectively. The rabbit phosphor-cugWT1 (S64/S68) antibodies were produced using a synthetic peptides (Ab Frontier). The pCMV6-cugWT1 plasmid was purchased from OriGene and the gene was recloned into the pcDNA3.1-V5 vector.

Lentiviral infection

The lentiviral expression vectors and packing vectors (pMD2.0G and psPAX) were purchased from Open Biosystem. The sequences were as follows: shGSK3β (#1, TRCN0000000822; 5′-CCCAAATGTCAAACTACCAAA-3′, #2, TRCN0000000823; 5′-CCGATTGCGTTATTTCTTCTA-3′, #3, TRCN0000000824; 5′-CCAATGTTTCGTATATCTGTT-3′, #4, TRCN0000039998; 5′-CCACTGATTATACCTCTAGTA-3′). The shRNAs were purchased from University of Minnesota Genomic Center and constructed using the protocol shown on the Open Biosystem website.

Cell culture and transfection

All cells were purchased from ATCC. Cells were cultured at 37°C in a 5% CO2 incubator following established ATCC protocols. Cells were confirmed to be mycoplasma negative. Vials of each cell line were thawed and maintained for up to 20 passages. HeLa and HEK293T cells were grown in DMEM/10% FBS. A549 cells were cultured with F-12K/10% FBS and MRC5 human normal lung fibroblasts were maintained in Eagle's Minimum Essential Medium/10% FBS. HCT15 cells was cultured in RPMI1640 medium/10% FBS. When cells reached 60% to 80% confluence, transfections were performed using the iMFectin poly DNA transfection reagent (GenDepot) according to the manufacturer's instructions.

Kinase assay

The GST fusion protein containing the N-terminal fragment of cugWT1 was produced in BL21 (DE3) Escherichia coli and the GST-C-terminal of the WT1 recombinant protein (Novus Biologicals) were incubated with recombinant active GSK3β (Cell Signaling Technology), 50 μmol/L unlabeled ATP and 10 μCi [γ-32P] ATP for 1 hour. After stopping the reaction with 6× SDS, samples were separated by SDS-PAGE, and phosphorylation was visualized by autoradiography.

Docking model

The GSK3β crystal structure was obtained from the Protein Data Bank (24) and prepared under standard procedures of the Protein Preparation Wizard (Schrödinger Suite 2019). The structure of the WT1 fragment (residues 34–100) was built based on multiple-threading alignments by local meta-threading-server and iterative threading assembly refinement (TASSER) of the protein structure and function predictions program I-TASSER 2.0 (25). GSK3β and WT1 protein–protein docking was performed by using 3-D Fast Fourier Transform–based protein docking algorithm of HEX 8.00, and we selected the best configuration to represent the binding mode. GSK3β and cugWT1 docking was demonstrated by the protein docking server with the interactive molecular graphics program (26). The best configuration to represent the docking model was selected for further analysis.

LC/MS-MS analysis to identify cugWT1 phosphorylation sites

The cugWT1 was expressed with or without constitutively active GSK3β into HEK293T cells. Cell lysates were immunoprecipitated with anti-Flag. WT1 proteins were eluted using urea buffer (7 M urea, 2 M thiourea, 2% 3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate) and digested with Glu-C (Promega) for 16 hours at 37°C. The Sciex TripleTOF TM 5600 system (ABSciex) was used to identify cugWT1 phosphorylation sites. The raw data were processed and searched with ProteinPilot software (version 4.5) using the Paragon algorithm. Protein identification was obtained by searching the UniProtKB human database, and filtered at ≥95% confidence cut off. Peptides for phosphorylated WT1 were identified at 5% global FDR level.

Animals and carcinogen treatment

These studies were approved by the University of Minnesota Institutional Animal Care and Use Committee. B6.129P2-Wt1tm1Hst/H mice were purchased from Infrafrontier. The mice were maintained under virus- and antigen-free conditions. Mice were genotyped by PCR analysis according to the Infrafrontier genotyping protocol. The primer sequences were: WT1 (F) 5′-GGT AAG GAG TTC AAG GCA GC-3′ and WT1 (WTR) 5′-CTG CGA TCC TGC GGA TTC T-3′. Male and female mice (6–8 weeks old) were randomly assigned to 6 groups: (i) WT-vehicle; (ii) WT-chemical; (iii) heterozygous-vehicle; (iv) heterozygous—chemical; (v) homozygous—vehicle; and (vi) homozygous—chemical.

Study 1: Colon carcinogenesis model

The AOM/DSS-induced model was established as described previously (27). Colon tumor growth was induced by azoxymethane (AOM; Sigma)/dextran sodium sulfate Mw40000 (DSS; BOC Sciences) AOM/DSS. Mice were injected intraperitoneally with 10 mg/kg body weight of AOM. One week later, 3% DSS was given in the drinking water for 1 week, followed by 2 weeks of regular water. The DSS cycle was repeated three times. To compare tumor incidence multiplicity, mice were euthanized after 14 weeks from first AOM injection by using CO2 (protocol ID: 1807-36161A).

Study 2: Lung carcinogenesis model

Lung tumor growth was induced by urethane (1 g/kg body weight, once a week for 10 weeks i.p.; Sigma; ref. 28). To compare tumor incidence multiplicity, mice were euthanized after 37 weeks from the first urethane injection, by using CO2 (protocol ID: 1709–35106A).

Tumors macroscopically visible on the colon or lung were counted and tissues harvested for further analysis. The tissues were fixed in 10% formalin and embedded in paraffin blocks. Tissues were sectioned with a microtome for subsequent hematoxylin and eosin (H&E) staining.

Additional methods

Western blotting, qRT-PCR, immunoprecipitation, and IHC were performed using standard techniques. Additional details are shown in Supplementary Materials and Methods.

Statistical analysis

Data are expressed as mean values SD form three independent experiments. Statistical analysis of data was performed using a Mann–Whitney U test or one-way ANOVA followed by Dunnett or Tukey test to determine significant differences. A P-value <0.05 was considered statistically significant.

CUG-WT1 is critical for AOM/DSS-induced colon carcinogenesis

To determine whether the absence of cugWT1 could halt or inhibit tumor growth or progression, we prepared WT1 WT (+/+), heterozygous (±), and homozygous (-/-) mice and confirmed that cugWT1 expression in homozygous was depleted but did not affect augWT1 production (Supplementary Figs. S1A–S1D). First, we treated mice with AOM/DSS (Fig. 1A), an established protocol of colon carcinogenesis that induces mutations by the mutagenic agent AOM and inflammation produced by DSS treatment, thus recapitulating the traits of human colon cancer pathogenesis (29). All the AOM/DSS-stimulated mice developed colon tumors, and we found that colon tumorigenesis in cugWT1 knockout mice was lower than in WT mice. We also observed an increase in colonic length with a reduction in intestinal shrinkage due to tumor growth and decreased tumor number (Fig. 1BD). In addition, the histologic analysis of colon after H&E staining showed that AOM/DSS-treated WT mice revealed extensive chronic inflammation, with tumors characterized by atypical epithelial cells with dysplastic nuclei. In contrast, less inflammation and mild dysplasia of colon was observed in cugWT1-deficient mice (Fig. 1E). A statistically significant inhibition was demonstrated in the increase level of PCNA, which is marker for evaluating cell proliferation, induced by AOM/DSS treatment in knockout mice as found by Western blotting (Fig. 1F). Notably the expression of cyclin D1 and Bcl2, which are cugWT1 target genes, was detected and analyzed by immunoblotting. Results indicated that cugWT1 deficient mice showed lower levels of these proteins compared with WT mice (Fig. 1G). Although the AOM/DSS-treated mice showed weight loss with severe diarrhea compared with untreated mice, no difference was observed between each genotyped group (Supplementary Fig. S1E). In contrast, the survival rate of the cugWT1 knockout AOM/DSS-treated group tended to be longer compared with the WT group (Supplementary Fig. S1G). Overall, these results demonstrated that cugWT1 promotes tumor development and progression in AOM/DSS-induced colon carcinogenesis.

Figure 1.

The cugWT1 protein is critical in the AOM/DSS-induced colon carcinogenesis model. A, Illustration of the experimental proceures used to induce colon carcinogenesis in B6.129P2-Wt1tm1Hst/H mice. Mice were injected intraperitoneally with a single dose (10 mg/kg body weight) of AOM. One week later, 3% DSS was administered in the drinking water for 1 week, followed by 2 weeks of regular water. The cycle of DSS treatment was repeated three times. Mouse colons were collected at the indicated end points. B, Images of colons at necropsy. C and D, Colon length (C) and number (D) of tumors in each group. WT (+/+)-vehicle-treated (n = 22); WT-AOM/DSS-treated (n = 17); heterozygous (±)-vehicle-treated (n = 21); heterozygous-AOM/DSS-treated (n = 21); homozygous (−/−)-vehicle-treated (n = 22); homozygous-AOM/DSS-treated (n = 22). E, H&E staining of tumor morphology in colon tissue (×100). Scale bar, 100 μm. F, PCNA expression level visualized by immunoblotting. Vehicle groups (n = 5 per group); AOM/DSS-treated group (n = 11 per each). G, expression levels of cyclin D1 and Bcl2, which are targets of WT1, as analyzed by immunoblotting. Vehicle groups (n = 5 per each) WT-AOM/DSS-treated (n = 14); heterozygous-AOM/DSS-treated (n = 20); homozygous-chemical-treated (n = 22). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 vs. WT; ##, P < 0.01 vs. vehicle, as analyzed by one-way ANOVA with Tukey correction.

Figure 1.

The cugWT1 protein is critical in the AOM/DSS-induced colon carcinogenesis model. A, Illustration of the experimental proceures used to induce colon carcinogenesis in B6.129P2-Wt1tm1Hst/H mice. Mice were injected intraperitoneally with a single dose (10 mg/kg body weight) of AOM. One week later, 3% DSS was administered in the drinking water for 1 week, followed by 2 weeks of regular water. The cycle of DSS treatment was repeated three times. Mouse colons were collected at the indicated end points. B, Images of colons at necropsy. C and D, Colon length (C) and number (D) of tumors in each group. WT (+/+)-vehicle-treated (n = 22); WT-AOM/DSS-treated (n = 17); heterozygous (±)-vehicle-treated (n = 21); heterozygous-AOM/DSS-treated (n = 21); homozygous (−/−)-vehicle-treated (n = 22); homozygous-AOM/DSS-treated (n = 22). E, H&E staining of tumor morphology in colon tissue (×100). Scale bar, 100 μm. F, PCNA expression level visualized by immunoblotting. Vehicle groups (n = 5 per group); AOM/DSS-treated group (n = 11 per each). G, expression levels of cyclin D1 and Bcl2, which are targets of WT1, as analyzed by immunoblotting. Vehicle groups (n = 5 per each) WT-AOM/DSS-treated (n = 14); heterozygous-AOM/DSS-treated (n = 20); homozygous-chemical-treated (n = 22). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 vs. WT; ##, P < 0.01 vs. vehicle, as analyzed by one-way ANOVA with Tukey correction.

Close modal

CUG-WT1 is critical for urethane-induced lung carcinogenesis

To further study cugWT1's role in carcinogenesis, we used a urethane-induced lung carcinogenesis model comparing the mice with WT or cugWT1-depleted mice. Urethane is well known as chemical carcinogen used to induce lung cancer as a model for human lung adenocarcinoma (30). We treated mice with urethane (Fig. 2A). The results indicated that lung tumors occurrence was dramatically decreased in cugWT1-deficient mice compared with WT at 37 weeks after starting urethane administration (Fig. 2B and C). Moreover, Kaplan–Meier survival curves revealed that the survival rate of the cugWT1 knockout urethane-treated group was significantly greater compared with the WT group (Fig. 2D). In addition, histologic and immunoblotting analysis showed that cugWT1-deficient mice reduced tumor formation and exhibited lower expression of PCNA and cyclin D1 compared with WT mice (Fig. 2EG). Notably, cugWT1-deficient mice exhibited expression levels of Bcl2 regardless of whether treated or not treated with urethane (Fig. 2G). Body weight was not different between each groups (Supplementary Fig. S1F). Collectively, these results indicated that inhibition of cugWT1 suppressed tumor development and progression in urethane-induced lung carcinogenesis and agreed with the results of the colon carcinogenesis study.

Figure 2.

The cugWT1 protein is critical for urethane-induced lung carcinogenesis. A, Experimental procedures used to induce lung carcinogenesis in B6.129P2-Wt1tm1Hst/H mice. Mice were injected intraperitoneally with urethane (1 g/kg body weight) once a week for 10 weeks. Mouse lungs were collected at the indicated end point. B, Images of lungs at necropsy. Arrows, tumor. C, Number of tumors in each group. WT (+/+)-urethane-treated (n = 16); heterozygous (±)-urethane-treated (n = 24); homozygous (−/−)-urethane-treated (n = 24). D, Kaplan–Meier survival curves (*, P < 0.05 by log-rank test). E, H&E staining of tumor morphology in lung tissue. Scale bar, 100 μm. F, PCNA expression level as assessed by immunoblotting. Vehicle groups (n = 5 per group); urethane-treated groups (n = 16 per group). G, Immunoblotting for cyclin D1 and Bcl2, which are targets of WT1 expression level. Vehicle groups (n = 5 per group); urethane-treated groups (n = 10 per group). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 vs. WT; ##, P < 0.01 vs. vehicle, as analyzed by one-way ANOVA with Tukey's correction.

Figure 2.

The cugWT1 protein is critical for urethane-induced lung carcinogenesis. A, Experimental procedures used to induce lung carcinogenesis in B6.129P2-Wt1tm1Hst/H mice. Mice were injected intraperitoneally with urethane (1 g/kg body weight) once a week for 10 weeks. Mouse lungs were collected at the indicated end point. B, Images of lungs at necropsy. Arrows, tumor. C, Number of tumors in each group. WT (+/+)-urethane-treated (n = 16); heterozygous (±)-urethane-treated (n = 24); homozygous (−/−)-urethane-treated (n = 24). D, Kaplan–Meier survival curves (*, P < 0.05 by log-rank test). E, H&E staining of tumor morphology in lung tissue. Scale bar, 100 μm. F, PCNA expression level as assessed by immunoblotting. Vehicle groups (n = 5 per group); urethane-treated groups (n = 16 per group). G, Immunoblotting for cyclin D1 and Bcl2, which are targets of WT1 expression level. Vehicle groups (n = 5 per group); urethane-treated groups (n = 10 per group). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 vs. WT; ##, P < 0.01 vs. vehicle, as analyzed by one-way ANOVA with Tukey's correction.

Close modal

CUG-WT1 phosphorylation of S64 stimulates its turnover and inhibits cell proliferation

In a previous study, we found that phosphorylation of cugWT1 at serine 68 protected cugWT1 from proteasomal degradation (13). To explore additional cugWT1 phosphorylation sites that might be associated with its degradation, we searched for phosphorylation sites within the 73 aa region and focused on S64 (Fig. 3A). We then determined whether mutation of this phosphorylation site would alter the half-life of cugWT1 proteins in HeLa cells by using cycloheximide chase assay. Results (Fig. 3B) indicated that the phosphor-mimetic mutation (S64E) appeared to induce more rapid turnover of cugWT1 compared with WT. The cugWT1 protein has previously been demonstrated to be ubiquitinated in cells, which promoted its degradation and shortened half-life (13). Thus, we examined the ubiquitination level in the active mutant forms of cugWT1. Results indicated that the S64E mutants had increased ubiquitination compared with WT cugWT1 (Fig. 3C). To investigate the contribution of S64 phosphorylation to cugWT1 activity, we measured proliferation in MRC5 normal cells. MRC5 is ideal for WT and S64 mutant cugWT1 transfections because of the absence of endogenous cugWT1. The data showed that the cugWT1-S64E-transfected cells exhibited reduced growth compared with WT (Fig. 3D). These data supported the idea that phosphorylation of cugWT1 S64 stimulates its turnover and inhibits cell proliferation. However, we also found that cugWT1 S64 is less active compared with cugWT1 S68 in lung and colon cancer cells or tissues (Supplementary Figs. S2A and S2B).

Figure 3.

The cugWT1 protein's phosphorylation at S64 stimulates its turnover and inhibits cell proliferation. A, Sequence of the putative relationship with cugWT1 degradation of phosphorylation sites in humans and mice. We focused on S64. B, Cycloheximide (CHX) chase to detect turnover of cugWT1 proteins. HeLa cells transfected with a V5-cugWT1 wt or active mutant (S64E) plasmid were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cells were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and subjected to Western blotting. GAPDH was used as a loading control. C, Ubiquitination of cugWT1. HeLa cells were transfected with a V5-cugWT1 or S64E plasmid along with HA-Ub and then treated with MG132 (10 μmol/L) for 6 hours. V5- or HA-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized with anti-HA. D, MRC5 cells expressing cugWT1 wt or S64E were plated at equal concentrations. Cell proliferation was monitored for 6 days using a live-cell imaging system (IncuCyte; Essen BioScience). Endogenous phosphorylated cugWT1 (S64) was detected by Western blotting. Results are shown as mean values ± SD. *, P < 0.05 by Mann-Whitney U test; n = 3.

Figure 3.

The cugWT1 protein's phosphorylation at S64 stimulates its turnover and inhibits cell proliferation. A, Sequence of the putative relationship with cugWT1 degradation of phosphorylation sites in humans and mice. We focused on S64. B, Cycloheximide (CHX) chase to detect turnover of cugWT1 proteins. HeLa cells transfected with a V5-cugWT1 wt or active mutant (S64E) plasmid were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cells were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and subjected to Western blotting. GAPDH was used as a loading control. C, Ubiquitination of cugWT1. HeLa cells were transfected with a V5-cugWT1 or S64E plasmid along with HA-Ub and then treated with MG132 (10 μmol/L) for 6 hours. V5- or HA-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized with anti-HA. D, MRC5 cells expressing cugWT1 wt or S64E were plated at equal concentrations. Cell proliferation was monitored for 6 days using a live-cell imaging system (IncuCyte; Essen BioScience). Endogenous phosphorylated cugWT1 (S64) was detected by Western blotting. Results are shown as mean values ± SD. *, P < 0.05 by Mann-Whitney U test; n = 3.

Close modal

CUG-WT1 is phosphorylated on S64 by GSK3β

In a search for potential kinases that can phosphorylate cugWT1 S64, we screened candidates by using a kinase substrate motif and found that cugWT1 S64 might be a GSK3β target site (31). To determine whether GSK3β phosphrylates cugWT1 S64, we performed kinase assays with active GSK3β and GST-WT1 N-terminal (1–73 aa), N-terminal S64A, or C-terminal (74–519 aa, augWT1). We detected phosphorylation in the WT1 N-terminal; however, no phosphorylation was observed in the N-terminal S64A or C-terminal (Fig. 4A). To confirm the phosphorylation in cells, we transfected cugWT1 and GSK3β into 293T cells and analyzed phosphopeptides of cugWT1 by using mass spectrometry (Fig. 4B). Next, to further confirm that cugWT1 S64 phosphorylation occurs in cells, we transfected HeLa cells with cugWT1 WT or S64A and GSK3β (S9A). An increase in phosphorylation of WT cugWT1 was observed in GSK3β (S9A) overexpressing cells; however, this effect disappeared in cugWT1 S64A-transfected cells (Fig. 4C). We next determined whether GSK3β was the dominant kinase for this site in HeLa cells by using shRNA depletion followed by examination of S64 phosphorylation (Fig. 4D). To explore whether cugWT1 interacts with GSK3β, immunoprecipitation was performed using extracts of HeLa cells overexpressing V5-cugWT1 and HA- GSK3β. We successfully coprecipitated cugWT1/GSK3β complexes by using either epitope (V5 or HA) tags (Fig. 4E). Importantly, we could show IP of basal endogenous cugWT1 and GSK3β present in A549 cells (Fig. 4F). A docking model suggested that GSK3β would mainly bind to the N-terminal of cugWT1. In addition, the cugWT1 S64 site is close to the ATP binding pocket (Fig. 4G). Overall, these results support the idea that GSK3β binds cugWT1 and phosphorylates S64.

Figure 4.

The cugWT1 protein is phosphorylated at S64 by GSK3β. A, Top, structures of glutathione S transferase (GST)-WT1-N, the N-terminal (73 aa) of cugWT1 or GST-WT1-C, and the remaining portion of cugWT1 that falls into augWT1 are shown. The GST-WT1 fragments were incubated with active GSK3β and 10 μCi [γ-32P] of ATP for 1 hour. Bottom, WT1 phosphorylation was visualized by autoradiography. B, The cugWT1 protein was coexpressed with constitutively active GSK3β into HEK293T cells. Cell lysates were immunoprecipitated with anti-Flag and analyzed by mass spectrometry to identify the cugWT1 phosphorylation sites. C, Lysates from HeLa cells expressing V5-cugWT1 wt or inactive mutant (S64A) together with HA-GSK3β (S9A) were analyzed for cugWT1 S64 phosphorylation by Western blotting. D, HeLa cells were treated with shRNAs to deplete GSK3β for 24 hours prior to transfection with cugWT1 wt or the S64A mutant. Cell lysates were analyzed for cugWT1 S64 phosphorylation by Western blotting using a phosphor-WT1 (S64) antibody. E, Extracts from HeLa cells transfected with V5-cugWT1 and HA-GSK3β (S9A) were immunoprecipitated with anti-V5. Western blotting revealed coprecipitated HA-GSK3β (S9A) or V5-cugWT1. F, Extracts from A549 cells were pretreated with MG132 (10 μmol/L) for 4 hours and immunoprecipitated with IgG, anti-GSK3β, or anti-WT1 (D8I7F) antibodies to reveal endogenous complexes. G, Modeling of GSK3β binding with WT1. GSK3β and WT1 are colored blue and red, respectively.

Figure 4.

The cugWT1 protein is phosphorylated at S64 by GSK3β. A, Top, structures of glutathione S transferase (GST)-WT1-N, the N-terminal (73 aa) of cugWT1 or GST-WT1-C, and the remaining portion of cugWT1 that falls into augWT1 are shown. The GST-WT1 fragments were incubated with active GSK3β and 10 μCi [γ-32P] of ATP for 1 hour. Bottom, WT1 phosphorylation was visualized by autoradiography. B, The cugWT1 protein was coexpressed with constitutively active GSK3β into HEK293T cells. Cell lysates were immunoprecipitated with anti-Flag and analyzed by mass spectrometry to identify the cugWT1 phosphorylation sites. C, Lysates from HeLa cells expressing V5-cugWT1 wt or inactive mutant (S64A) together with HA-GSK3β (S9A) were analyzed for cugWT1 S64 phosphorylation by Western blotting. D, HeLa cells were treated with shRNAs to deplete GSK3β for 24 hours prior to transfection with cugWT1 wt or the S64A mutant. Cell lysates were analyzed for cugWT1 S64 phosphorylation by Western blotting using a phosphor-WT1 (S64) antibody. E, Extracts from HeLa cells transfected with V5-cugWT1 and HA-GSK3β (S9A) were immunoprecipitated with anti-V5. Western blotting revealed coprecipitated HA-GSK3β (S9A) or V5-cugWT1. F, Extracts from A549 cells were pretreated with MG132 (10 μmol/L) for 4 hours and immunoprecipitated with IgG, anti-GSK3β, or anti-WT1 (D8I7F) antibodies to reveal endogenous complexes. G, Modeling of GSK3β binding with WT1. GSK3β and WT1 are colored blue and red, respectively.

Close modal

CUG-WT1 expression is mediated by GSK3β

Notably, we found that GSK3β phosphorylates cugWT1 S64 and phosphorylation of cugWT1 S64 increased its degradation. Therefore, we next assessed the effects of GSK3β mediation on the estimated expression of cugWT1 proteins. The results showed that treatment with LiCl, the most common GSK3β inhibitor (32), or knockdown of GSK3β increased cugWT1 expression (Fig. 5A and B). In contrast, treatment with LY294002, a GSK3β indirect activator, or overexpression of GSK3β decreased cugWT1 protein levels (Fig. 5C and D). In addition, we confirmed that cugWT1 expression with GSK3β overexpression in HCT15 (colon) and A549 (lung) cancer cells reduced cugWT1 protein levels; and EGFR, cyclin D1, and Bcl2 expression was also decreased (Fig. 5E). These proteins are known to be critical oncogenic drivers and are associated with high cell proliferation (33–35). Suppression of cell proliferation by transfection of GSK3β was also observed in those cells (Fig. 5F).

Figure 5.

The cugWT1 protein expression is regulated by GSK3β. A, A549 cells were treated with LiCl as a GSK3β inhibitor for 0 and 12 hours. Cell extracts were used for Western blotting to detect WT1 (6F-H2) and phosphor-GSK3β (S9) expression. B, HeLa cells were treated for 24 hours with shRNAs to deplete GSK3β. Cell lysates were used to detect WT1 (6F-H2) by Western blotting. C, A549 cells were treated for 24 hours with LY294002 (10 μmol/L) as a GSK3β activator. Cell extracts were used for Western blotting to detect WT1 (6F-H2) expression. D, A cycloheximide (CHX) chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt with or without HA-GSK3β (S9A) and then treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and protein visualized by Western blotting. E, HCT15 and A549 cells were transfected with HA-GSK3β (S9A) and WT1 expression level was confirmed and EGFR, cyclin D1 and Bcl2, which are targets of WT1, expression was assessed. F, HCT15 and A549 cells expressing GSK3β (S9A) were plated at equal concentrations. Cell proliferation was monitored for 4 days using a live-cell imaging system (IncuCyte; Essen BioScience). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 by Mann-Whitney U test; n = 3.

Figure 5.

The cugWT1 protein expression is regulated by GSK3β. A, A549 cells were treated with LiCl as a GSK3β inhibitor for 0 and 12 hours. Cell extracts were used for Western blotting to detect WT1 (6F-H2) and phosphor-GSK3β (S9) expression. B, HeLa cells were treated for 24 hours with shRNAs to deplete GSK3β. Cell lysates were used to detect WT1 (6F-H2) by Western blotting. C, A549 cells were treated for 24 hours with LY294002 (10 μmol/L) as a GSK3β activator. Cell extracts were used for Western blotting to detect WT1 (6F-H2) expression. D, A cycloheximide (CHX) chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt with or without HA-GSK3β (S9A) and then treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and protein visualized by Western blotting. E, HCT15 and A549 cells were transfected with HA-GSK3β (S9A) and WT1 expression level was confirmed and EGFR, cyclin D1 and Bcl2, which are targets of WT1, expression was assessed. F, HCT15 and A549 cells expressing GSK3β (S9A) were plated at equal concentrations. Cell proliferation was monitored for 4 days using a live-cell imaging system (IncuCyte; Essen BioScience). Results are shown as mean values ± SD. *, P < 0.05; **, P < 0.01 by Mann-Whitney U test; n = 3.

Close modal

CUG-WT1 turnover and cell proliferation mediated by GSK3β is dependent on phosphorylation of S64

Phosphorylation by GSK3β is known to impact the ubiquitination and degradation of some of its substrates (17). On the basis of our previous results, we hypothesized that GSK3β might mediate cugWT1 turnover and cell proliferation through its phosphorylation of WT1 on S64. Thus, we compared WT1 stability between cugWT1 wt and S64A mutant proteins in HeLa with or without GSK3β expression by using cycloheximide chase assays. Results showed that GSK3β stimulates cugWT1 turnover in cells transfected with cugWT1 wt; however, this effect disappeared in S64A inactivating mutant cells (Fig. 6A). In addition, cugWT1 ubiquitination increased in cells cotransfected with wt and GSK3β, but not with S64A and GSK3β (Fig. 6B). MRC5 cells coexpressing with cugWT1 wt and GSK3β exhibited decreased proliferation, and EGFR and Bcl2 mRNA levels. However, this effect disappeared in the S64A inactivating mutant cells (Fig. 6C and D). These results suggested that GSK3β promotes cugWT1 turnover and inhibites cell proliferation through its phosphorylation of WT1 on S64.

Figure 6.

The cugWT1 protein turnover and cell proliferation induced by GSK3β depend on phosphorylation of WT1 at S64. A, A cycloheximide (CHX) chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt or S64A with or without HA-GSK3β (S9A) and then were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and proteins visualized by Western blotting. B, Ubiquitination of cugWT1. HeLa cells were transfected with V5-cugWT1 wt or S64A with or without HA-GSK3β (S9A) and Flag-Ub and were treated with MG132 (10 μmol/L) for 6 hours. V5-, HA-, or Flag-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized by Western blotting with anti-Flag. C, MRC5 cells expressing cugWT1 wt or S64A with or without HA-GSK3β (S9A) were plated at equal concentrations. Cell proliferation was monitored for 6 days using a live-cell imaging system (IncuCyte; Essen BioScience). D, EGFR and Bcl2 mRNA levels were detected by using quantitative real-time RT-PCR. Results are shown as mean values ± SD. *, P < 0.05 and **, P < 0.01, vs. without GSK3β (S9A) by one-way ANOVA with Tukey correction, n = 3.

Figure 6.

The cugWT1 protein turnover and cell proliferation induced by GSK3β depend on phosphorylation of WT1 at S64. A, A cycloheximide (CHX) chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt or S64A with or without HA-GSK3β (S9A) and then were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and proteins visualized by Western blotting. B, Ubiquitination of cugWT1. HeLa cells were transfected with V5-cugWT1 wt or S64A with or without HA-GSK3β (S9A) and Flag-Ub and were treated with MG132 (10 μmol/L) for 6 hours. V5-, HA-, or Flag-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized by Western blotting with anti-Flag. C, MRC5 cells expressing cugWT1 wt or S64A with or without HA-GSK3β (S9A) were plated at equal concentrations. Cell proliferation was monitored for 6 days using a live-cell imaging system (IncuCyte; Essen BioScience). D, EGFR and Bcl2 mRNA levels were detected by using quantitative real-time RT-PCR. Results are shown as mean values ± SD. *, P < 0.05 and **, P < 0.01, vs. without GSK3β (S9A) by one-way ANOVA with Tukey correction, n = 3.

Close modal

CUG-WT1 turnover associated with FBXW8 is dependent on phosphorylation of S64

Many GSK3β substrats are targeted for phosphorylation-mediated degradation in concert with E3 ubiquitination ligase. FBXWs are a family of E3 ligases, and we tested the interaction of cugWT1 with various isoforms of FBXWs. We found that FBXW4, FBXW5, and FBXW8 could bind to cugWT1 (Fig. 7A). Further, cugWT1 S64A decreased the interaction with FBXW8, but had no effect with the interaction with FBXW4 or FBXW5 was observed (Fig. 7B). Previously, we reported that FBXW8 induced ubiquitination of cugWT1 and in this additional study we showed that ubiquitination of the WT1 S64A mutant was not affected (Fig. 7C). Moreover, WT1 stability was increased by FBXW8 in the S64A inactive mutant. In contrast, WT1 stability was decreased by FBXW8 in the S64E active mutation (Fig. 7D). However no difference between wt and mutant cugWT1 with FBXW4 and FBXW5 was observed (Supplementary Fig. S3). Collectively, these data demonstrated that FBXW8 induced cugWT1 turnover when it was phosphorylated on S64.

Figure 7.

The cugWT1 protein turnover induced by FBXW8 depends on phosphorylation of S64. A, extracts from HeLa cells were transfected with V5-cugWT1 and Flag-FBXWs (FBXW2, FBXW4, FBXW5, FBXW7, and FBXW8) and were immunoprecipitated with anti-V5. Western blots revealed coprecipitated V5-cugWT1or Flag-FBXWs. B, Extracts from HeLa cells transfected with V5-cugWT1 or S64A and Flag-FBXWs (FBXW4, FBXW5, and FBXW8) were immunoprecipitated with anti-V5. Western blotting revealed coprecipitated V5-cugWT1or Flag-FBXWs. C, Ubiquitination of cugWT1. HeLa cells were transfected with V5-cugWT1 or S64A with Flag-FBXWs (FBXW5 or FBXW8) and HA-Ub were treated with MG132 (10 μmol/L) for 6 hours. V5-, HA-, and Flag-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized with anti-HA. D, A cycloheximide chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt or mutants (S64A, S64E) together with Flag-FBXW8 and were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and visualized by Western blotting. E, Schematic summarizing the mechanism cugWT1 degradation.

Figure 7.

The cugWT1 protein turnover induced by FBXW8 depends on phosphorylation of S64. A, extracts from HeLa cells were transfected with V5-cugWT1 and Flag-FBXWs (FBXW2, FBXW4, FBXW5, FBXW7, and FBXW8) and were immunoprecipitated with anti-V5. Western blots revealed coprecipitated V5-cugWT1or Flag-FBXWs. B, Extracts from HeLa cells transfected with V5-cugWT1 or S64A and Flag-FBXWs (FBXW4, FBXW5, and FBXW8) were immunoprecipitated with anti-V5. Western blotting revealed coprecipitated V5-cugWT1or Flag-FBXWs. C, Ubiquitination of cugWT1. HeLa cells were transfected with V5-cugWT1 or S64A with Flag-FBXWs (FBXW5 or FBXW8) and HA-Ub were treated with MG132 (10 μmol/L) for 6 hours. V5-, HA-, and Flag-tagged proteins were immunoprecipitated with anti-V5. Ubiquitinated cugWT1 proteins were visualized with anti-HA. D, A cycloheximide chase assay was used to detect turnover of cugWT1 proteins. HeLa cells were transfected with V5-cugWT1 wt or mutants (S64A, S64E) together with Flag-FBXW8 and were treated with cycloheximide (20 μg/mL) to prevent further protein synthesis. Whole cell lysates were harvested and prepared at 0, 3, and 6 hours after cycloheximide treatment and visualized by Western blotting. E, Schematic summarizing the mechanism cugWT1 degradation.

Close modal

In this study, we report an oncogenic role for CUG-translated WT1 (cugWT1) in two chemically induced cancer models in mice that are deficient in cugWT1. Results indicate that cugWT1 might be a potential target for cancer therapy and prevention.

WT1 is well-known as a chameleon gene, acting both as a tumor suppressor and an oncogene under different but unclear conditions. WT1 protein isoforms arise from canonical AUG-translated WT1 (augWT1) and CUG-translated WT1 (cugWT1; ref. 11). Recent ribosome profiling studies suggest that many translation initiation sites utilize non-AUG initiation codons (e.g., CUG, GUG), which are one nucleotide different from AUG (36, 37). If one or more non-AUG initiation codons are located upstream of the AUG initiation codon, the transcript can encode multiple protein isoforms that start with an alternative non-AUG initiation codon or a novel protein (38, 39). This alternative initiation has been observed in PTEN, HCK, and c-Myc and can generate isoforms with alternative localization properties and N-terminal extensions with modified activity (40–42). In the case of the oncogene c-Myc, the two isoforms have altered DNA-binding capacity, and the ratio of their expression levels regulates their activity (41). Our group has clearly shown that cugWT1 is overexpressed in various types of cancer cells, including lung and colon, and provided strong evidence showing that cugWT1 acts as an oncogene, whereas, augWT1 functions as a tumor suppressor (13). In addition, cugWT1 is also highly expressed in human cancer tissues. In this study, we aimed to expand on the functional importance of cugWT1. We used two different chemically-induced tumor mouse models: (i) a colitis-associated colorectal cancer induced by AOM/DSS and (ii) a non–small cell lung tumor model induced by urethane. Both models used WT and cugWT1 deficient mice. The knockout mice were produced as previously reported by introducing a translation stop signal downstream of the CUG initiator (12). The cugWT1 protein is absent in this mutant mouse but the translation from the AUG start codon was not affected (43). The lack of cugWT1 was confirmed in mutant embryos and cancer tissues (Supplementary Figs. S1B, S1C, and S1F). Our results showed that the development of chemically-induced tumors was dramatically depressed in cugWT1-mutant mice. In addition, the findings also revealed decreased expression of cyclin D1 and Bcl2 in mutant mice, which supported an oncogenic role of cugWT1 in vivo (Figs. 1 and 2). However, why WT1 has different translation processes like augWT1 and cugWT1, and why tumor cells undergo such selective translation processes, remains unclear.

We also found that GSK3β promotes cugWT1 phosphorylation at S64, resulting in ubiquitination and degradation of cugWT1 in concert with FBXW8 (Fig. 7E).

The cugWT1protein contains potential phosphorylation sites located within the extra amino-acid sequence of the N-terminal extension, which is not present in the shorter augWT1 (43). Our group first discovered that phosphorylation of cugWT1 at S68 results in resistance to ubiquitination, resulting in reduced degradation (13). In this study, we found that phosphorylation of cugWT1 on S64 stimulated its turnover and inhibited cell proliferation (Fig. 3).

GSK3β is a proline-directed serine/threonine kinase and the most common target for phosphorylation by GSK3β is the sequence, S/T-X-X-X-S/T(P). GSK3β phosphorylates proteins at sites in their serine/threonine N-terminal (44) and in this case, it phosphorylates the cugWT1 at S64RGAS68 in its N-terminal. Furthermore, results of a kinase assay, LC/MS-MS, and GSK3β overexpression or knockout clearly showed that GSK3β binds to and phosphorylates cugWT1 on S64 (Fig. 4). GSK3β mediates cellular metabolism and gene expression (45) and also cell survival by phosphorylating pro-proliferative factors for degradation. C-Myc is an example of proteins with short half-lives that are phosphorylated by GSK3β, resulting in ubiquitination and degradation through the proteasome degradation pathway (46). Similarly, cugWT1 is a target of GSK3β and its expression is controlled by GSK3β. As a transcription factor, WT1 preserves the survival of cancer cells through its modulation of several important genes associated with proliferation. For example, WT1 increases the expression of EGFR (13), cyclin D1 (2), and Bcl2 (3), and therefore, inhibiting WT1 should at least indirectly mediate these genes. Consistent with these reports, we found that phosphorylation by GSK3β decreased the expression of cugWT1 followed by the downregulation of EGFR, cyclin D1 and Bcl2 expression levels in HCT15 and A549 cancer cells (Fig. 5). Crosstalk between phosphorylation and the ubiquitin machinery is important for regulating protein activity, stability, and interactions (47, 48). Our results using an inactive mutant cugWT1- S64A indicated that GSK3β promotes cugWT1 turnover through its phosphorylation on S64 (Fig. 6).

Ubiquitin is conjugated to target proteins through the E1 (ubiquitin-activating enzyme), E2 (ubiquitin-conjugating enzyme), and E3 (ubiquitin ligase) activities (49). The F-box/WD repeat-containing proteins (FBXW) constitute one of the subunits of ubiquitin-protein ligase (E3) complexes, which function in phosphorylation-dependent ubiquitination. The FBXWs also comprise one of the four subunits of the ubiquitin ligase complex, SCFs (SKP1-cullin-F-box), which function in phosphorylation-dependent ubiquitination. Our results showed that only FBXW8 recognized the cugWT1 protein phosphorylation S64 (Fig. 7), suggesting that GSK3β cooperates with FBXW8 for phosphorylation-mediated degradation.

In cancer cells or tissues, whereas strong Akt activity was observed, we detected very little active GSK3β probably because GSK3β is phosphorylated on Ser9 by Akt, resulting in its inactivation. Consistent with our data, WT1 S64 is less active than WT1 S68 in cancer cells or tissues (Supplementary Figs. S2A and S2B). Thus, development of GSK3β activators that target cancer cells resulting in inhibition of oncogenic cugWT1 might offer a logical strategy for chemoprevention or treatment. Importantly, GSK3β would target cugWT1, but not augWT1, because S64 is only present in cugWT1. However, questions about the specificity of GSK3β still remain unanswered. No other known protains appear to be responsible for S64 phosphorylation, but will be a topic for future studies.

In conclusion, our studies demonstrated an oncogenic role of CUG-translated WT1 (cugWT1) in vivo. Inhibition of cugWT1 expression provides a potential candidate target for therapy. Moreover, we reported that GSK3β promotes cugWT1 phosphorylation at S64, resulting in ubiquitination and degradation of cugWT1 in concert with FBXW8. This novel mechanism for forcing turnover of cugWT1 might be used for new treatment.

No disclosures were reported.

H. Yoshitomi: Conceptualization, data curation, formal analysis, methodology, writing–original draft. K.Y. Lee: Conceptualization, data curation. K. Yao: Conceptualization, methodology, writing–review and editing. S.H. Shin: Conceptualization, methodology. T. Zhang: Data curation, methodology. Q. Wang: Methodology. S. Paul: Methodology. E. Roh: Methodology. J. Ryu: Data curation. H. Chen: Data curation. F. Aziz: Methodology. A. Chakraborty: Methodology. A.M. Bode: Writing–review and editing. Z. Dong: Conceptualization, supervision, writing–review and editing.

The authors thank Tara Adams for supporting animal experiments. This work was supported by the Karl R. Potach Foundation and The Hormel Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Call
KM
,
Glaser
T
,
Ito
CY
,
Buckler
AJ
,
Pelletier
J
,
Haber
DA
, et al
Isolation and characterization of a zinc finger polypeptide gene at the human chromosome 11 Wilms' tumor locus
.
Cell
1990
;
60
:
509
20
.
2.
Xu
C
,
Wu
C
,
Xia
Y
,
Zhong
Z
,
Liu
X
,
Xu
J
, et al
WT1 promotes cell proliferation in non-small cell lung cancer cell lines through up-regulating cyclin D1 and p-pRb in vitro and in vivo
.
PLoS One
2013
;
8
:
e68837
.
3.
Loeb
DM
. 
WT1 influences apoptosis through transcriptional regulation of Bcl-2 family members
.
Cell Cycle
2006
;
5
:
1249
53
.
4.
Oji
Y
,
Miyoshi
S
,
Maeda
H
,
Hayashi
S
,
Tamaki
H
,
Nakatsuka
S
, et al
Overexpression of the Wilms' tumor gene WT1 in de novo lung cancers
.
Int J Cancer
2002
;
100
:
297
303
.
5.
Oji
Y
,
Yamamoto
H
,
Nomura
M
,
Nakano
Y
,
Ikeba
A
,
Nakatsuka
S
, et al
Overexpression of the Wilms' tumor gene WT1 in colorectal adenocarcinoma
.
Cancer Sci
2003
;
94
:
712
7
.
6.
Oji
Y
,
Yano
M
,
Nakano
Y
,
Abeno
S
,
Nakatsuka
S
,
Ikeba
A
, et al
Overexpression of the Wilms' tumor gene WT1 in esophageal cancer
.
Anticancer Res
2004
;
24
:
3103
8
.
7.
Loeb
DM
,
Evron
E
,
Patel
CB
,
Sharma
PM
,
Niranjan
B
,
Buluwela
L
, et al
Wilms' tumor suppressor gene (WT1) is expressed in primary breast tumors despite tumor-specific promoter methylation
.
Cancer Res
2001
;
61
:
921
5
.
8.
Miyoshi
Y
,
Ando
A
,
Egawa
C
,
Taguchi
T
,
Tamaki
Y
,
Tamaki
H
, et al
High expression of Wilms' tumor suppressor gene predicts poor prognosis in breast cancer patients
.
Clin Cancer Res
2002
;
8
:
1167
71
.
9.
Tatsumi
N
,
Oji
Y
,
Tsuji
N
,
Tsuda
A
,
Higashio
M
,
Aoyagi
S
, et al
Wilms' tumor gene WT1-shRNA as a potent apoptosis-inducing agent for solid tumors
.
Int J Oncol
2008
;
32
:
701
11
.
10.
Huff
V
. 
Wilms' tumours: about tumour suppressor genes, an oncogene and a chameleon gene
.
Nat Rev Cancer
2011
;
11
:
111
21
.
11.
Dallosso
AR
,
Hancock
AL
,
Brown
KW
,
Williams
AC
,
Jackson
S
,
Malik
K
. 
Genomic imprinting at the WT1 gene involves a novel coding transcript (AWT1) that shows deregulation in Wilms' tumours
.
Hum Mol Genet
2004
;
13
:
405
15
.
12.
Miles
CG
,
Slight
J
,
Spraggon
L
,
O'Sullivan
M
,
Patek
C
,
Hastie
ND
. 
Mice lacking the 68-amino-acid, mammal-specific N-terminal extension of WT1 develop normally and are fertile
.
Mol Cell Biol
2003
;
23
:
2608
13
.
13.
Lee
KY
,
Jeon
YJ
,
Kim
HG
,
Ryu
J
,
Lim
DY
,
Jung
SK
, et al
The CUG-translated WT1, not AUG-WT1, is an oncogene
.
Carcinogenesis
2017
;
38
:
1228
40
.
14.
Makki
MS
,
Heinzel
T
,
Englert
C
. 
TSA downregulates Wilms tumor gene 1 (Wt1) expression at multiple levels
.
Nucleic Acids Res
2008
;
36
:
4067
78
.
15.
Smolen
GA
,
Vassileva
MT
,
Wells
J
,
Matunis
MJ
,
Haber
DA
. 
SUMO-1 modification of the Wilms' tumor suppressor WT1
.
Cancer Res
2004
;
64
:
7846
51
.
16.
Ye
Y
,
Raychaudhuri
B
,
Gurney
A
,
Campbell
CE
,
Williams
BR
. 
Regulation of WT1 by phosphorylation: inhibition of DNA binding, alteration of transcriptional activity and cellular translocation
.
EMBO J
1996
;
15
:
5606
15
.
17.
Garvin
AJ
,
Khalaf
AHA
,
Rettino
A
,
Xicluna
J
,
Butler
L
,
Morris
JR
, et al
GSK3beta-SCFFBXW7alpha mediated phosphorylation and ubiquitination of IRF1 are required for its transcription-dependent turnover
.
Nucleic Acids Res
2019
;
47
:
4476
94
.
18.
Okabe
H
,
Lee
SH
,
Phuchareon
J
,
Albertson
DG
,
McCormick
F
,
Tetsu
O
. 
A critical role for FBXW8 and MAPK in cyclin D1 degradation and cancer cell proliferation
.
PLoS One
2006
;
1
:
e128
.
19.
Manoukian
AS
,
Woodgett
JR
. 
Role of glycogen synthase kinase-3 in cancer: regulation by Wnts and other signaling pathways
.
Adv Cancer Res
2002
;
84
:
203
29
.
20.
Mills
CN
,
Nowsheen
S
,
Bonner
JA
,
Yang
ES
. 
Emerging roles of glycogen synthase kinase 3 in the treatment of brain tumors
.
Front Mol Neurosci
2011
;
4
:
47
.
21.
Ougolkov
AV
,
Billadeau
DD
. 
Targeting GSK-3: a promising approach for cancer therapy?
Future Oncol
2006
;
2
:
91
100
.
22.
Jemal
A
,
Bray
F
,
Center
MM
,
Ferlay
J
,
Ward
E
,
Forman
D
. 
Global cancer statistics
.
CA Cancer J Clin
2011
;
61
:
69
90
.
23.
Weir
HK
,
Thun
MJ
,
Hankey
BF
,
Ries
LA
,
Howe
HL
,
Wingo
PA
, et al
Annual report to the nation on the status of cancer, 1975–2000, featuring the uses of surveillance data for cancer prevention and control
.
J Natl Cancer Inst
2003
;
95
:
1276
99
.
24.
Berman
HM
,
Westbrook
J
,
Feng
Z
,
Gilliland
G
,
Bhat
TN
,
Weissig
H
, et al
The protein data bank
.
Nucleic Acids Res
2000
;
28
:
235
42
.
25.
Yang
J
,
Yan
R
,
Roy
A
,
Xu
D
,
Poisson
J
,
Zhang
Y
. 
The I-TASSER Suite: protein structure and function prediction
.
Nat Methods
2015
;
12
:
7
8
.
26.
Macindoe
G
,
Mavridis
L
,
Venkatraman
V
,
Devignes
MD
,
Ritchie
DW
. 
HexServer: an FFT-based protein docking server powered by graphics processors
.
Nucleic Acids Res
2010
;
38
:
W445
9
.
27.
Wei
TT
,
Lin
YT
,
Chen
WS
,
Luo
P
,
Lin
YC
,
Shun
CT
, et al
Dual targeting of 3-hydroxy-3-methylglutaryl coenzyme A reductase and histone deacetylase as a therapy for colorectal cancer
.
EBioMedicine
2016
;
10
:
124
36
.
28.
Wang
Q
,
Gao
G
,
Zhang
T
,
Yao
K
,
Chen
H
,
Park
MH
, et al
TRAF1 is critical for regulating the BRAF/MEK/ERK pathway in non-small cell lung carcinogenesis
.
Cancer Res
2018
;
78
:
3982
94
.
29.
Tanaka
T
,
Kohno
H
,
Suzuki
R
,
Yamada
Y
,
Sugie
S
,
Mori
H
. 
A novel inflammation-related mouse colon carcinogenesis model induced by azoxymethane and dextran sodium sulfate
.
Cancer Sci
2003
;
94
:
965
73
.
30.
Xu
C
,
Zhou
L
,
Lu
L
,
Chen
T
,
Wei
S
,
Lin
X
, et al
Inflammation has a role in urethaneinduced lung cancer in C57BL/6J mice
.
Mol Med Rep
2016
;
14
:
3323
8
.
31.
Frame
S
,
Cohen
P
. 
GSK3 takes centre stage more than 20 years after its discovery
.
Biochem J
2001
;
359
:
1
16
.
32.
Ryves
WJ
,
Harwood
AJ
. 
Lithium inhibits glycogen synthase kinase-3 by competition for magnesium
.
Biochem Biophys Res Commun
2001
;
280
:
720
5
.
33.
Jimeno
A
,
Hidalgo
M
. 
Pharmacogenomics of epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors
.
Biochim Biophys Acta
2006
;
1766
:
217
29
.
34.
Radha
G
,
Raghavan
SC
. 
BCL2: A promising cancer therapeutic target
.
Biochim Biophys Acta Rev Cancer
2017
;
1868
:
309
14
.
35.
Semczuk
A
,
Jakowicki
JA
. 
Alterations of pRb1-cyclin D1-cdk4/6-p16(INK4A) pathway in endometrial carcinogenesis
.
Cancer Lett
2004
;
203
:
1
12
.
36.
Fritsch
C
,
Herrmann
A
,
Nothnagel
M
,
Szafranski
K
,
Huse
K
,
Schumann
F
, et al
Genome-wide search for novel human uORFs and N-terminal protein extensions using ribosomal footprinting
.
Genome Res
2012
;
22
:
2208
18
.
37.
Ingolia
NT
,
Lareau
LF
,
Weissman
JS
. 
Ribosome profiling of mouse embryonic stem cells reveals the complexity and dynamics of mammalian proteomes
.
Cell
2011
;
147
:
789
802
.
38.
Slavoff
SA
,
Mitchell
AJ
,
Schwaid
AG
,
Cabili
MN
,
Ma
J
,
Levin
JZ
, et al
Peptidomic discovery of short open reading frame-encoded peptides in human cells
.
Nat Chem Biol
2013
;
9
:
59
64
.
39.
Touriol
C
,
Bornes
S
,
Bonnal
S
,
Audigier
S
,
Prats
H
,
Prats
AC
, et al
Generation of protein isoform diversity by alternative initiation of translation at non-AUG codons
.
Biol Cell
2003
;
95
:
169
78
.
40.
Carreno
S
,
Caron
E
,
Cougoule
C
,
Emorine
LJ
,
Maridonneau-Parini
I
. 
p59Hck isoform induces F-actin reorganization to form protrusions of the plasma membrane in a Cdc42- and Rac-dependent manner
.
J Biol Chem
2002
;
277
:
21007
16
.
41.
Hann
SR
,
King
MW
,
Bentley
DL
,
Anderson
CW
,
Eisenman
RN
. 
A non-AUG translational initiation in c-myc exon 1 generates an N-terminally distinct protein whose synthesis is disrupted in Burkitt's lymphomas
.
Cell
1988
;
52
:
185
95
.
42.
Hopkins
BD
,
Fine
B
,
Steinbach
N
,
Dendy
M
,
Rapp
Z
,
Shaw
J
, et al
A secreted PTEN phosphatase that enters cells to alter signaling and survival
.
Science
2013
;
341
:
399
402
.
43.
Bruening
W
,
Pelletier
J
. 
A non-AUG translational initiation event generates novel WT1 isoforms
.
J Biol Chem
1996
;
271
:
8646
54
.
44.
Beurel
E
,
Grieco
SF
,
Jope
RS
. 
Glycogen synthase kinase-3 (GSK3): regulation, actions, and diseases
.
Pharmacol Ther
2015
;
148
:
114
31
.
45.
Woodgett
JR
. 
Judging a protein by more than its name: GSK-3
.
Sci STKE
2001
;
2001
:
re12
.
46.
Welcker
M
,
Orian
A
,
Jin
J
,
Grim
JE
,
Harper
JW
,
Eisenman
RN
, et al
The Fbw7 tumor suppressor regulates glycogen synthase kinase 3 phosphorylation-dependent c-Myc protein degradation
.
Proc Natl Acad Sci U S A
2004
;
101
:
9085
90
.
47.
Hunter
T
. 
The age of crosstalk: phosphorylation, ubiquitination, and beyond
.
Mol Cell
2007
;
28
:
730
8
.
48.
Zhao
Y
,
Brickner
JR
,
Majid
MC
,
Mosammaparast
N
. 
Crosstalk between ubiquitin and other post-translational modifications on chromatin during double-strand break repair
.
Trends Cell Biol
2014
;
24
:
426
34
.
49.
Liu
J
,
Shaik
S
,
Dai
X
,
Wu
Q
,
Zhou
X
,
Wang
Z
, et al
Targeting the ubiquitin pathway for cancer treatment
.
Biochim Biophys Acta
2015
;
1855
:
50
60
.

Supplementary data