O‐GlcNAcylation promotes colorectal cancer progression by regulating protein stability and potential catcinogenic function of DDX5

Abstract The RNA helicase p68 (DDX5), a key player in RNA metabolism, belongs to the DEAD box family and is involved in the development of colorectal cancer. Here, we found both DDX5 and O‐GlcNAcylation are up‐regulated in colorectal cancer. In addition, DDX5 protein level is significantly positively correlated with the expression of O‐GlcNAcylation. Although it was known DDX5 protein could be regulated by post‐translational modification (PTM), how O‐GlcNAcylation modification regulated of DDX5 remains unclear. Here we show that DDX5 interacts directly with OGT in the SW480 cell line, which is the only known enzyme that catalyses O‐GlcNAcylation in humans. Meanwhile, O‐GlcNAcylation could promote DDX5 protein stability. The OGT‐DDX5 axis affects colorectal cancer progression mainly by regulating activation of the AKT/mTOR signalling pathway. Taken together, these results indicated that OGT‐mediated O‐GlcNAcylation stabilizes DDX5, promoting activation of the AKT/mTOR signalling pathway, thus accelerating colorectal cancer progression. This study not only reveals the novel functional of O‐GlcNAcylation in regulating DDX5, but also reveals the carcinogenic effect of the OGT‐DDX5 axis in colorectal cancer.

significantly elevated in gastric cancer and co-excited the mTOR signalling pathway to enhance the growth of gastric cancer cells.
Protein O-GlcNAcylation is a broad and dynamic modification of β-N-acetyl-D-glucosamine. [9][10][11][12] It is a specific type of post-translational modification catalysed by O-linked N-acetylglucosamine transferase (OGT), resulting in the transfer of O-linked β-N-acetylglucosamine (O-GlcNAcylation) to the serine of the target protein or threonine residue. 9,13 Previous studies have shown that it can induce conformational changes to initiate protein folding, compete with phosphorylation of the same or proximal serine or threonine, and disrupt protein-protein interactions. And regulate protein stability. [14][15][16][17] Pathologically, abnormal O-GlcNAcylation has been shown to stimulate tumorigenesis in various cancers by modulating cell signalling, transcription, cell division, metabolism, and cytoskeletal regulation. 12,[18][19][20] Colorectal cancer (CRC) is one of the major health problems with high mortality worldwide. 21 Epidemiological studies have shown a strong association between O-GlcNAcylation and poor prognosis of CRC. 22 In CRC patients, the synthesis of UDP-N-acetyl-Dglucosamine (UDP-GlcNAc) is increased, which is the substrate for OGT for O-GlcNAcylation of target proteins and O-GlcNAcylation of certain proteins may be a key factor in tumorigenesis. 23 In addition, abnormal activation of AKT/mTOR signalling is the key factor of CRC, which leads to tumour growth and metastasis. [24][25][26] Previous studies have shown that DDX5 can transcriptionally co-activate AKT signaling pathway in CRC to regulate cancer progression. 7 Because of the significant increase in the prevalence of CRC, it is important to understand the molecular basis underlying this disease and the potential role of DDX5 and O-GlcNAcylation in CRC.
In this study, we demonstrated that O-GlcNAcylation of DDX5 is involved in colorectal cancer progression by activation of AKT/mTOR signalling pathway, which opens up a new sight for the treatment of colorectal cancer.

| Cell lines and cell culture
Human CRC cell lines HT29, HCT116, SW480, SW620, and normal intestinal epithelial cells NCM460 were obtained from the Chinese Academy of Sciences Cell Bank (China). NCM460, HT29, HCT116, SW480, and SW620 cells were cultured in DMEM medium supplemented with 10% (v/v) FBS (Gibco) and 1% penicillin/streptomycin. All cells were cultured at 37°C with 5% (v/v) CO 2 . All cells were tested for mycoplasma contamination.

| RNA extraction, reverse transcription, and real-time RT-PCR
Total RNA from cells was extracted using an RNA isolation kit (Qiagen, Germany) according to the manufacturer's instructions.
Subsequently, the RevertAid First Strand cDNA Synthesis Kit (TaKaRa, Japan) was used to reverse-transcribe the messenger RNA (mRNA) from the total mRNA; the specific primer (Table S1) and the SYBR premix Ex Taq (TaKaRa) were used to expand by realtime qPCR. It was carried out with the following parameters: predenaturation at 95°C for 5 minutes, denaturation at 95°C for 10 seconds, annealing at 62°C for 20 seconds, and extension at 72°C for 30 seconds for 40 cycles. GAPDH was used as an internal control.

| Cycloheximide or Thiamet-G treatment
Cycloheximide (Sigma-Aldrich, USA) was dissolved in dimethyl sulfoxide (Sigma-Aldrich) to a concentration of 500 mg/mL and diluted in DMEM to a final concentration of 50 mg/mL. Thiamet-G (Calbiochem, San Diego, USA) was dissolved in DMSO to a concentration of 40 mmol/L and diluted to a final concentration of 10 μmol/L in DMEM medium. Cells were incubated with DMEM medium containing 10% FBS and 1% penicillin/streptomycin to achieve a cell concentration of 80% prior to cycloheximide or Thiamet-G treatment. Next, cells were treated with cycloheximide or Thiamet-G (10 μmol/L) in the absence of FBS for 12 hours.

| Western blotting
For Western blot (WB), the protein was resolved on SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gel followed by standard WB.

| Co-immunoprecipitation
The cells were collected and lysed with IP lysis buffer (Thermo Scientific, #87787) for 1 hour. Cell lysates were collected and incubated overnight with PureProteome Protein A or Protein G Magnetic Beads (Millipore, #LSKMAGA02) and antibodies at 4°C; immunoprecipitates were washed five times and then subjected to immunoblot analysis.

| Cell viability assay
The cells were seeded in a 96-well plate (1 × 10 4 cells/well), and cultured in a 37°C, 5% CO 2 humidified incubator for 24 hours. Ten microlitres of Cell Counting Kit-8 solution (Dojindo, Kumanoto, Japan) was added to each well and incubated for 2 hours at 37°C in a 5% CO 2 humidified incubator. Spectrometer Varioskan ® Flash (Thermo Fisher, Waltham, USA) was used to measure absorbance at 450 nm. A proliferation curve is drawn in which time is taken as the WU ET AL.
| 1355 abscissa and the average absorbance value in each group is taken as the ordinate. The experiment was performed in triplicate.

| Colony formation assay
The cells were seeded in 6-well plates (500 cells/well) and incubated at 37°C in a 5% CO 2 humidified incubator. After 2 weeks, the cells were stained with Gentian violet (Beyotime Biotechnology, Shanghai, China). The experiment was repeated three times.
Patient information provided by the TMA supplier. After the tissue sections were deparaffinized and rehydrated, antigen retrieval was carried out in citrate buffer (Beyotime) at 100°C for 0.5 hour.
Endogenous peroxidase was blocked with 3% peroxide for 15 min-

| mTOR agonist and inhibitor treatment
Rapamycin (Santa Cruz, sc-3504) and MHY1485 (Sigma, SML0810) were dissolved in dimethyl sulfoxide (DMSO) as a stock solution at 1 and 5 mM, respectively. The cells were cultured in DMEM medium containing 10% FBS. When the number of cells reached 80%, cells were treated with rapamycin (50 nM) and MHY1485 (10 nM) for 12 hours. The author's sequence is available upon request.

| Statistical analysis
Tests used to examine differences between groups included Student's t test, one-way and two-way ANOVA, and Chi-squared test. P < 0.05 was considered statistically significant.

| Correlation of DDX5 with O-GlcNAcylation in vitro and in vivo
To explore the correlation between DDX5 and O-GlcNAcylation expression, tissue microarray (TMA) of human CRC tissue was used.  Figure 1I and J). In addition, we found that knockdown of OGT resulted in a significant decrease in DDX5 levels ( Figure 1K). Conversely, overexpression of OGT increases the level of DDX5, indicating that there may be a correlation between DDX5 and OGT ( Figure 1K). In summary, our results indicate that DDX5 and O-GlcNAcylation may be correlated in vivo and in vitro. tumour-promoting function. 27 We found that knockdown of OGT resulted in a significant decrease in DDX5 and O-GlcNAcylation levels ( Figure 2A) in SW480 cells, as well as the abilities of cell proliferation ( Figure 2B), colony formation ( Figure 2D), and migration ( Figure 2E). Interestingly, we found that overexpression of DDX5 on the basis of OGT knockdown could partly rescue the abilities of cell proliferation ( Figure 2B), colony formation ( Figure 2D), and migration ( Figure 2E). Meanwhile, knockdown of DDX5 can significantly block the effect of promoting cell proliferation which mediated by overexpression of OGT ( Figure 2C). We also found that xenografts grew much more slowly in OGT knockdown group than control group, and overexpression of DDX5 also partly reversed the growth of xenografts ( Figure 2F-H). Collectively, these in vitro and in vivo results indicated that stimulation of cellular O-GlcNAcylation can promote colorectal tumorigenesis in a DDX5-dependent manner.

| O-GlcNAc modification increase DDX5 protein stability
We subsequently investigated how O-GlcNAcylation regulates DDX5 expression levels. Firstly, we found that O-GlcNAcylation did not affect DDX5 mRNA levels. (Figure S1A). Since O-GlcNAcylation is reported to be an important factor affecting protein stability, 12 we therefore investigated whether stimulation of O-GlcNAcylation inhibits DDX5 degradation. To clarify whether down-regulation of O-GlcNAcylation alters the stability of DDX5, SW480 cells with or without OGT knockdown were treated with cycloheximide to inhibit protein synthesis, and then the remaining DDX5 was detected.
As shown in Figure 3A, the protein level of DDX5 was reduced to 50% of its original amount in the cells of the control group at 18.5 hours after the cycloheximide treatment. However, in cells with OGT knockdown, DDX5 was reduced to 50% of its original amount at 7.5 hours after the cycloheximide treatment. Similar results showed that DDX5 protein levels were reduced to 50% at
These results indicate that OGT regulates the stability of DDX5 mainly by modifying the DDX5 protein through O-GlcNAcylation and increasing protein stability.

| OGT interacts with DDX5 in CRC cell line
O-GlcNAcylation is well known to be dynamically regulated by OGT and OGA. 22 To elucidate the exact mechanism by which

| OGT-DDX5 axis affects colorectal cancer proliferation and metastasis by regulating AKT/mTOR pathway
Previous studies have found that DDX5 can regulate the expression of AKT and phosphorylation in colorectal cancer, thus affecting the development of colorectal cancer. 7 It has also been reported that DDX5 can regulate the phosphorylation activity of mTOR and S6K1 in gastric cancer, thereby affecting the progression of gastric cancer. 4 We found that knockdown of OGT significantly inhibited AKT/mTOR signalling pathway activity ( Figure 4A). Conversely, overexpression of OGT activates the AKT/mTOR signalling pathway ( Figure 4B). Meanwhile, we found that knockdown DDX5 also significantly inhibited AKT/mTOR signal activation ( Figure 4C). Similarly, overexpression of DDX5 activates the AKT/mTOR signalling pathway ( Figure 4D). In summary, we propose to envision whether the OGT-DDX5 axis regulates the proliferation and metastasis of colorectal cancer by activating the AKT/mTOR pathway. To demonstrate that the OGT-DDX5 axis regulates colorectal cancer cell proliferation and metastasis by activating the AKT/mTOR pathway. We overexpressed DDX5 or added an mTOR agonist (MHY1485) in the OGT knockdown colorectal cancer cell line. The results indicated that the cell proliferation and metastasis ability of DDX5 overexpressing and mTOR agonists is significantly enhanced in OGT knockdown SW480 ( Figure 4E, F). Similarly, we knocked down DDX5 on the basis of OGT overexpression or used the mTOR inhibitor (Rapamycin) significantly reduce the proliferation and metastasis of colorectal cancer cells ( Figure 4G, H).
Therefore, we found that OGT knockdown can significantly inhibit the expression of DDX5, thereby changing the phosphorylation activity of AKT and mTOR ( Figure 4A). At the same time, we found that overexpression of DDX5 in OGT knockdown cells significantly reversed the inhibition of the AKT/mTOR pathway ( Figure 4A).  Figure 4B). We also found that it significantly inhibited the activity of AKT/mTOR signalling pathway by knocking down DDX5 in the OGT overexpressing cell lines ( Figure 4B).
To further verify that the OGT-DDX5 axis regulates colorectal cancer progression through the AKT/mTOR pathway. We knocked down DDX5 in SW480 and found that it can significantly inhibit colorectal cancer proliferation ( Figure S1B). However, the addition of the mTOR agonist (MHY1485) significantly reversed cell proliferation ( Figure S1B). Similarly, we added a mTOR inhibitor (Rapamycin) to DDX5 overexpressing cell lines to reduce cell proliferation changes caused by DDX5 overexpression ( Figure S1C). Notably, we found that AKT/mTOR activity was significantly reduced in DDX5 knock- in cancer tissues.
A recent study showed that DDX5 is significantly higher in colorectal cancer than normal tissues, and DDX5 is involved in phosphorylation of AKT and mTOR. 7 It has also been reported that activated mTOR signalling leads to increased expression of OGT and O-GlcNAcylation in colorectal cancer. 14 Figure S1D). Therefore, we detected the stable binding products of OGT and DDX5 by co-IP. And the level of O-GlcNAcylation affects the stability of DDX5 protein.
There is growing evidence that it is significantly associated with tumorigenesis. 37,38 Recent studies have found that DDX5 stimulates the development of gastric cancer by activating mTOR signalling. 4 Cross-regulated colorectal tumorigenesis between mTOR signalling and O-GlcNAcylation was recently reported. 14 Interestingly, in the current study, there was no report that mTOR was modified by O-GlcNAcylation. Therefore, we speculate that the OGT regulation of mTOR can be achieved in other ways. However, the way in which OGT regulates mTOR remains unclear and further research is needed.
Here, we prove that DDX5 can be O