Hepatocyte-specific Wtap deficiency promotes hepatocellular carcinoma by activating GRB2–ERK depending on downregulation of proteasome-related genes

Wilm’s tumor 1–associating protein (WTAP), a regulatory protein of the m6A methyltransferase complex, has been found to play a role in regulating various physiological and pathological processes. However, the in vivo role of WTAP in the pathogenesis of hepatocellular carcinoma (HCC) is unknown. In this study, we have elucidated the crucial role of WTAP in HCC progression and shown that hepatic deletion of Wtap promotes HCC pathogenesis through activation of multiple signaling pathways. A single dose of diethylnitrosamine injection causes more and larger HCCs in hepatocyte-specific Wtap knockout (Wtap-HKO) mice than Wtapflox/flox mice fed with either normal chow diet or a high-fat diet. Elevated CD36, IGFBP1 (insulin-like growth factor–binding protein 1), and chemokine (C–C motif) ligand 2 (CCL2) expression leads to steatosis and inflammation in the Wtap-HKO livers. The hepatocyte proliferation is dramatically increased in Wtap-HKO mice, which is due to higher activation of extracellular signal–regulated kinase (ERK) and signal transducer and activator of transcription-3 signaling pathways. Hepatic deletion of Wtap activates the ERK signaling pathway by increasing the protein stability of GRB2 and ERK1/2, which is due to the decreased expression of proteasome-related genes. Restoring PSMB4 or PSMB6 (two key components of the proteasome) leads to the downregulation of GRB2 and ERK1/2 in Wtap-HKO hepatocytes. Mechanistically, WTAP interacts with RNA polymerase II and H3K9ac to maintain expression of proteasome-related genes. These results demonstrate that hepatic deletion of Wtap promotes HCC progression through activating GRB2–ERK1/2-mediated signaling pathway depending on the downregulation of proteasome-related genes especially Psmb4 and Psmb6.

Hepatocellular carcinoma (HCC) is a common hepatobiliary cancer with high rates of morbidity and mortality.The development and progression of HCC depend on the intricate interplay of genetic, environmental, and lifestyle factors.Most cases of HCC occur because of chronic liver damage caused by exposure to carcinogens, viral hepatitis infections, alcohol consumption, and nonalcoholic steatohepatitis (NASH) (1).The development of HCC involves various factors, such as viruses, growth factors, metabolic molecules, and inflammatory cytokines (2).These factors can activate signaling pathways like Ras-Raf-MEK (mitogen-activated protein kinase kinase)-extracellular signal-regulated kinase (ERK), MEKK1/ c-Jun N-terminal kinase (JNK), PI3K/Akt/mTOR, Wnt/bcatenin, Janus kinase-signal transducer and activator of transcription-3 (STAT3), insulin-like growth factor (IGF) receptor, hepatocyte growth factor receptor (HGFR), and epidermal growth factor receptor (EGFR) that can result in the initiation, invasion, and metastasis of HCC (3)(4)(5).However, how these signaling pathways are integratively regulated is not well known.Understanding the integrative molecular mechanisms involved in the progression of HCC in response to environmental stimuli can aid in identifying early diagnostic markers or new targets for drug development.
Many disease processes such as NASH, obesity, and diabetes have been found to be regulated by m 6 A RNA writer proteins such as methyltransferase like 3 (METTL3) and Wilm's tumor 1-associating protein (WTAP) by altering m 6 A RNA modification or chromatin accessibility (6)(7)(8)(9)(10)(11). METTL3 and WTAP have been found to be abnormally elevated in HCC (12,13), and knocking down METTL3 or WTAP in HCC cell lines limited cell proliferation (13,14).As a result, researchers concluded that METTL3 and WTAP increased HCC development, and that inhibiting METTL3 or WTAP may be a viable approach for HCC treatment (13,14).However, HCC cell lines and nude animal models may not correctly reflect the activity of METTL3 and WTAP in HCC formation in vivo.Recently, it was demonstrated that hepatocyte-specific deletion of Mettl3 promotes HCC in mice by activating numerous signaling pathways (15).However, the role of WTAP in HCC development in vivo is mainly unknown.It is also unknown if WTAP affects signaling pathways related to HCC in an integrative manner.
In this study, we used a mouse model with hepatocytespecific Wtap knockout (Wtap-HKO) to investigate the role of WTAP in HCC progression.We have demonstrated that hepatic deletion of Wtap promotes the pathogenesis of HCC by activating multiple signaling pathways, at least partially through stabilizing GRB2 and ERK1/2 protein levels.

Hepatocyte-specific deletion of Wtap dramatically induces HCC in mice
To test if WTAP influences HCC pathogenesis in vivo, we delivered a single intraperitoneal injection of diethylnitrosamine (DEN) (50 mg/kg) to male Wtap-HKO and Wtap flox/flox mice at 2 weeks old and evaluated liver tumors at 35 weeks.Wtap-HKO mice had a bigger liver weight and more and larger HCCs than Wtap flox/flox mice (Fig. 1, A-D).Cell proliferation in Wtap-HKO livers was considerably boosted, as evidenced by an increase in Ki67-positive cells (Fig. 1E).These findings show that deleting Wtap in the liver enhances HCC development.

Wtap deletion in hepatocytes promotes HCC progression in young mice
A recent study found that WTAP controls nonalcoholic fatty liver disease and NASH progression, and Wtap-HKO mice showed NASH-like phenotypes in both normal chow (NC) and NASH diet-fed circumstances (7).We wanted to see if Wtap-HKO mice develop HCC under short-term feeding circumstances.After a short period of feeding (11 weeks) of either NC or high-fat diet (HFD) following DEN injection, Wtap-HKO mice developed considerably more and larger HCC tumors, whereas Wtap flox/flox mice did not (Fig. 2, A-C).Cell proliferation was consistently increased in both NC-and HFD-fed DEN-treated Wtap-HKO mice, as seen by increased Ki67-positive cells (Fig. 2, D and E).These findings show that hepatic Wtap deletion exacerbates DEN-induced HCC development under short-term feeding circumstances.
We then investigated whether hepatic Wtap deletion accelerates DEN-induced HCC progression by enhancing liver steatosis and inflammation.Following DEN treatment, Wtap-HKO mice showed higher liver triglyceride levels than Wtap flox/flox mice under both NC diet (35 weeks) and HFD (11 weeks) feeding circumstances (Fig. 2, F and G), which is likely related to elevated serum-free fatty acid (Fig. 2H) and expression of CD36 and IGFBP1 (insulin-like growth factorbinding protein 1) (Fig. 2, I and J) (7).A tumor-promoting inflammatory microenvironment is also critical in the development of HCC.The expression of important cytochemokines such as chemokine (C-C motif) ligand 2 (CCL2) and tumor necrosis factor alpha (TNFα) was shown to be dramatically enhanced in Wtap-HKO mice (Fig. 2, I-K) (7).Some members of the interleukin (IL) 6 cytokine family, such as CLCF1 and LIF, were elevated, but others, such as CNTF, CTF1, IL6, IL11, IL27, and OSM, were not (Fig. 2K).Elevated CLCF1 and LIF levels may help activate STAT3 (16,17).STAT3 phosphorylation was dramatically increased in Wtap-HKO mice (Figs.2L and S1), contributing to higher cell proliferation.The classical TNF signaling cascade (phosphorylation of IκBα and p65) was not elevated in Wtap-HKO mice treated with or without DEN (Figs. 2L and S1).

Wtap deletion in hepatocytes activates the GRB2-ERK signaling pathway
In addition to the elevated STAT3 signaling pathway, Wtap-HKO mice have considerably increased ERK phosphorylation (Fig. 3, A, B, C, E, F, and G), which contributes to higher cell proliferation.Increased GRB2 and p-MEK1/2 levels contributed to ERK activation (Fig. 3, A, B, C, E, F, and  G).Wtap-HKO livers had higher amounts of ERK1/2 protein (Fig. 3, A, B, C, E, F, and G), which may contribute to ERK activation.However, upstream receptor activation (EGFR and HGFR/Met) was intact or decreased (Fig. S2, A-F), showing that EGFR and HGFR/Met are unlikely to contribute to GRB2-MEK-ERK signaling pathway activation in Wtap-HKO livers.Gab1 and p-B-Raf levels were modestly elevated in Wtap-HKO livers, which may also contribute to ERK signaling pathway activation (Fig. S2, A-E).GRB2, p-MEK1/ 2, p-ERK1/2, and ERK1/2 levels were likewise increased in isolated Wtap-HKO hepatocytes (Fig. 3, D and H).These findings show that hepatic Wtap deletion exacerbates DENinduced HCC formation, at least in part, via activating the GRB2-ERK signaling pathway.Hepatocyte-specific deficiency of Wtap increases protein stability of GRB2 and ERK1/2 We investigated how Wtap deficiency in hepatocytes raises GRB2 and ERK1/2 protein levels.GRB2 and ERK1/2 protein levels were observed to be elevated in the livers and isolated hepatocytes of Wtap-HKO mice (Fig. 3, A-H).Only Erk1 mRNA levels were modestly elevated, whereas Grb2 and Erk2 mRNA levels were not (Fig. 4, A-C), implying that Wtap deficiency promotes translational efficiency or protein stability.We measured translational efficiency in Wtap flox/flox and Wtap-HKO mice by sucrose gradient centrifugation of Wtap flox/flox and Wtap-HKO polysome fractions, and RT-PCR of endogenous mRNA levels of Grb2, Erk1, and Erk2 in different ribosome fractions.As shown in Figure 4, D-H, the relative distribution of Grb2, Erk1, and Erk2 mRNAs in polysome fractions was not altered in Wtap-HKO livers, showing that hepatic deletion of Wtap had no effect on the translational efficiency of Grb2, Erk1, and Erk2 mRNA.GRB2 and ERK1/2 protein stability was also assessed.GRB2 and ERK1/2 protein stability was considerably elevated in isolated hepatocytes from Wtap-HKO mice (Fig. 4,  I-K).These findings show that hepatic deficiency of Wtap increases the protein stability of GRB2 and ERK1/2.

Hepatocyte-specific deficiency of Wtap reduces the expression of proteasome-related genes, which contributes to higher protein stability of GRB2 and ERK
To determine how hepatic deficiency of Wtap increases protein stability of GRB2 and ERK1/2, we analyzed our previous RNA-Seq data (GSE168850) (7) and noticed that downregulated genes were associated with the proteasome (Fig. 5A), and 28 proteasomeassociated genes were downregulated (Fig. 5B), which was confirmed by RT-quantitative PCR (qPCR) (Fig. S3).Proteasome activity was significantly decreased in Wtap-HKO livers (Fig. 5C).This was also observed in Mettl3-HKO livers (18).Among the downregulated proteasome components, PSMB4 and PSMB6 were involved in METTL3-mediated MKK4-JNK protein stability (18).PSMB4 and PSMB6 protein levels were also considerably reduced in Wtap-HKO livers or primary hepatocytes (Fig. 5, D-F).Furthermore, MG132, an inhibitor of the proteasome, raised GRB2 and ERK1/2 protein levels in primary hepatocytes (Fig. 5, G and H), demonstrating that decreased proteasome activity leads to higher GRB2 and ERK1/2 protein levels.
Restoring proteasome activity via overexpression of PSMB4 or PSMB6 decreases GRB2 and ERK1/2 protein levels in Wtap-

HKO hepatocytes
We investigated whether restoring proteasome activity via PSMB4 or PSMB6 overexpression could correct the increased protein levels of GRB2 and ERK1/2 in Wtap-HKO hepatocytes.We isolated primary hepatocytes from Wtap flox/flox and Wtap-HKO mice, respectively.Hepatocytes were infected with an equal amount of Ad-βGal, Ad-FLAG-PSMB4, or Ad-FLAG-PSMB6 adenoviruses overnight.Proteasome activity was significantly decreased (Fig. 6A), whereas the GRB2 and ERK1/2 protein levels were upregulated in Wtap-HKO hepatocytes (Fig. 6, B-G).PSMB4 or PSMB6 overexpression restored proteasome activity (Fig. 6A), which reversed the upregulation of GRB2 and ERK1/2 in Wtap-HKO hepatocytes (Fig. 6, B-G).These findings suggest that hepatic deficiency of Wtap reduces the expression of proteasome-related genes especially Psmb4 and Psmb6, which contributes to the increased protein stability of GRB2 and ERK1/2.
Wtap deficiency promotes hepatocellular carcinoma METTL14-METTL3 has been demonstrated to influence gene expression via binding to RNA polymerase II and active histone modifications such as H3K36me3, H3K9ac, and H3K27ac (18,19).WTAP interacted with RNA polymerase II (Rpb1 C-terminal domain [CTD]) and H3K9ac in primary hepatocytes (Fig. 7I).The binding of RNA polymerase II (Rpb1 CTD) to genes of Psmb4 and Psmb6 was reduced in the livers of Wtap-HKO mice, whereas the binding of RNA polymerase  II (Rpb1 CTD) to Actb was unaffected (Fig. 7J).These findings suggest that WTAP is essential for maintaining the expression of proteasome components through interacting with RNA polymerase II and H3K9ac.
In Wtap-HKO livers, many signaling pathways contribute to the development of HCC.Hepatic deletion of Wtap induces lipolysis in white adipose tissue and NASH in the liver (7), exacerbating DEN-induced HCC development.IGFBP1 has been demonstrated to raise serum-free fatty acid by stimulating lipolysis in white adipose tissue.Wtap-HKO mice have abnormally raised CD36 levels, which contributes to the enhanced steatosis in both NC diet-and HFD-fed Wtap-HKO mice.Inflammation is another factor that contributes to HCC development.TNFα and CCL2 levels were considerably increased in Wtap-HKO mice (7), which promotes HCC growth.WTAP directly regulates the transcription of Igfbp1, Cd36, and Ccl2, which increases the development of NASH (7).Elevated CLCF1 and LIF levels may lead to STAT3 activation, which enhances tumor cell proliferation, including HCC (25).Hepatic deletion of Wtap does not increase the canonical TNF signaling cascade (phosphorylation of IκBα and p65), implying that WTAP modulates cytochemokine production independent of the TNF signaling pathway.  .WTAP binds to the promoters of proteasome-related genes and promotes their expression.A and B, primary hepatocytes were isolated from Wtap-HKO and Wtap flox/flox mice and then treated with ACTD (10 μg/ml) for different times (0, 3, 6, and 12 h).RT-qPCR was used to determine Psmb4 and Psmb6 mRNA levels (n = 6 independent cell samples for each group).C-F, sucrose gradient centrifugation was used to identify the fractions of 40s, 60s, 80s, and polysome in the livers of Wtap-HKO and Wtap flox/flox mice.Semiquantitative RT-PCR was used to determine the relative distribution of Psmb4 and Psmb6 mRNAs in these fractions (n = 3 mice for each group).G, a combined analysis of the ATAC-Seq (GSE168945), RNA-Seq (GSE168850), and ChIP-Seq (GSE198023) data revealed that WTAP binds to the promoters of 11 proteasome-related genes, including Psmb4, Psmb6, Psmb1, Psmb7, Psma3, Psmb3, Psma6, Psma5, Psmb2, Psma1, and Psme4, and their expression was downregulated in Wtap-HKO livers.H, ChIP-qPCR was used to assess FLAG-WTAP binding to the promoters of Psmb4, Psmb6, and Actb genes in the primary hepatocytes infected with Ad-FLAG-WTAP or βGal adenovirus.FLAG beads were used to immunoprecipitate the hepatocyte samples.Immunoprecipitated DNA was extracted for qPCR analysis (n = 4-6 for each group; Psmb4: p = 0.0302; Psmb6: p = 0.0022; Actb: p = 0.3862).I, primary hepatocytes were isolated from WT mice and then infected with βGal and FLAG-WTAP adenoviruses, respectively.Total cell lysates were treated with DNase1 (200 U/ml) at 37 C for 30 min.FLAG beads were used to immunoprecipitate the cell lysates, which were subsequently immunoblotted with anti-H3K9ac anti-Rpb1 CTD and anti-FLAG antibodies.This experiment was conducted twice independently with similar results.J, ChIP-qPCR was used to assess RNA polymerase II (Rpb1 CTD) levels binding to the promoters of Psmb4, Psmb6 The phosphorylation of ERK1/2 is significantly raised in both NC-and HFD-fed Wtap-HKO mice, in addition to IGFBP1, CD36, CCL2, and p-STAT3.High ERK signaling pathway activation promotes tumor cell growth, including HCC (25).ERK activation may be linked to upstream signaling molecules such as GRB2 and p-MEK1/2 being activated in Wtap-HKO livers.Wtap-HKO livers have considerably greater ERK1/2 protein levels, which may lead to higher p-ERK1/2 levels.The phosphorylation of EGFR and Met is not elevated in Wtap-HKO livers, indicating that WTAP regulates ERK signaling pathway activation independent of EGFR and Met.
Our investigation focused on how WTAP deficiency in hepatocytes affects GRB2 and ERK1/2 activation and expression.Hepatic deletion of Wtap does not affect translational efficiency of Grb2 and Erk1/2 mRNA, whereas Wtap-HKO livers have a lower overall global translational efficiency.Instead, protein stability of GRB2 and ERK1/2 is elevated in Wtap-HKO livers.RNA-Seq identifies multiple downregulated proteasome-related genes (including Psmb4 and Psmb6) in Wtap-HKO mice.GRB2 and ERK1/2 protein levels are increased when the proteasome is inhibited by MG132.Both PSMB4 and PSMB6 overexpression reduce GRB2 and ERK1/2 protein levels in Wtap-HKO hepatocytes.These findings suggest that deleting WTAP in the liver raises GRB2 and ERK1/2 protein levels through downregulating proteasome-related genes, particularly Psmb4 and Psmb6.Mettl3-HKO livers also have higher MKK4/JNK, GRB2, p-ERK, and ERK protein levels, which are likely attributable to downregulation of proteasome-related genes (15,18), demonstrating that WTAP and METTL3 share the same molecular processes.We noticed that downregulated proteasome activity in Mettl3-HKO and Wtap-HKO livers does not increase all the proteins.The detailed molecular mechanisms of the selection need further investigation.Ubiquitin-mediated proteolysis may be also involved in this process.
WTAP binds to METTL3 and acts as a regulator of the m 6 A RNA methyltransferase complex (12).Wtap deletion in hepatocytes significantly alters gene expression profiles (7).Some alterations are dependent on m 6 A modification, whereas others are dependent on chromatin accessibility that is independent of m 6 A modification (7).Although Psmb4 and Psmb6 are two of the m 6 A peak downregulated genes, their mRNA stability and translational efficiency remain unchanged.Instead, chromatin accessibility in the promoters of Psmb4 and Psmb6 is reduced.The binding of RNA polymerase II to the promoters of Psmb4 and Psmb6 is consistently reduced in Wtap-HKO livers.WTAP interacts with RNA polymerase II and H3K9ac on a molecular level.METTL14 and METTL3 have also been demonstrated to influence gene expression by binding to RNA polymerase II and H3K36me3 (19).METTL3 and WTAP have also been demonstrated to bind to gene promoters and influence gene transcription via distinct mechanisms (6,7,26).These findings suggest that WTAP is essential for maintaining the expression of proteasome components (including PSMB4 and PSMB6) through binding to RNA polymerase II and active histone modifications such as H3K9ac.
Loss of m 6 A writer protein WTAP or METTL3 in hepatocytes enhances hepatic steatosis, inflammation, and activation of STAT3 and ERK signaling pathways, which exacerbates the course of DEN-induced HCC.Wtap or Mettl3 deletion in other tissues or cell types, such as brown adipose tissue and islet β cells, results in tissue/cell injury and illness (6,(8)(9)(10)(11)18).These findings suggest that the m 6 A writer proteins WTAP and METTL3 are required for maintaining tissue homeostasis.On the other hand, WTAP and METTL3 are abundantly expressed in HCC and are necessary for tumor cell proliferation (13,14). Wap or Mettl3 knockdown in HCC cell lines reduces HCC cell proliferation, cell survival, migration, and colony formation, resulting in the inhibition of HCC tumorigenicity in nude mouse models (13,14).These findings further suggest that hepatic WTAP and METTL3 may play separate roles in HCC development.Downregulation or inactivation of WTAP or METTL3 during the initial stage, for example, may induce HCC, whereas upregulation or activation of WTAP or METTL3 during the later stage may accelerate HCC.This idea is supported by current data.Nuclear WATP or METTL3 downregulation has been linked to NASH (6,7), and hepatic Wtap or Mettl3 deletion enhances liver damage, nonalcoholic fatty liver disease, and HCC initiation (6,7,27), whereas WTAP and METTL3 are abnormally elevated in HCC (13,14), promoting HCC progression.
In conclusion, our findings show that hepatic Wtap deletion enhances HCC progression by activating GRB2-ERK1/2mediated signaling pathway dependent on the downregulation of proteasome-related genes, particularly Psmb4 and Psmb6.

Animal experiments
The Guide for the Care and Use of Laboratory Animals was strictly followed during animal research.The animal experiment protocols were approved by Institutional Animal Care and Use Committee (IACUC) of the Harbin Institute of Technology with authorization number IACUC-2018004.Mice were kept on a 12 h light/12 h dark cycle and fed various diets with free access to water.Wtap flox/flox mice have been previously described (7,9,11).Briefly, exon 4 of Wtap gene was flanked by two loxp sites using the CRISPR-Cas9 technique.Wtap flox/flox mice were crossed with Alb-Cre mice to generate Wtap-HKO mice.The genotype of Wtap-HKO mice is Wtap flox/flox Alb-Cre +/− .All these mice were in C57BL/ 6J background.Male Wtap-HKO mice and Wtap flox/flox control mice were given a single dose of DEN (50 mg/kg) intraperitoneally at 2 weeks old for DEN-induced HCC.The mice were randomly divided into several groups.Wtap-HKO and Wtap flox/flox mice were fed an NC diet for 35 weeks before being killed for HCC investigation.For 11 weeks, the second and third groups of mice were fed an NC diet and HFD, respectively.Ki67 immunostaining has previously been demonstrated (10,28).

Primary hepatocyte culture and adenoviral infection
Primary hepatocytes were isolated from C57BL/6 WT, Wtap flox/flox , and Wtap-HKO mice via liver perfusion with type II collagenase (Worthington Biochem) and cultured in RPMI1640 medium supplemented with 3% fetal bovine serum at 37 C in 5% CO 2 incubator.Cells were infected with an equal amount of Ad-βGal, Ad-FLAG-PSMB4, or Ad-FLAG-PSMB6 adenoviruses overnight.

Proteasome activity assay
Liver or hepatocyte samples were lysed in 50 mM Tris-HCl, pH 7.5, 250 mM sucrose, 5 mM MgCl 2 , 1 mM DTT, 2 mM ATP, 0.5 mM EDTA, and 0.025% digitonin for 5 min in ice.The lysate was cleared by centrifugation for 10 min at 12,000 rpm.Protein concentration was measured by bicinchoninic acid assay.Equal amounts of total protein were incubated in the assay buffer containing 50 mM Tris-HCl (pH 7.5), 5 mM MgCl 2 , 40 mM KCl, 1 mM DTT, 2 mM ATP, 50 μM Z-LLE-AMC (amino-4-methylcoumarin; for caspaselike activity) in the presence or the absence of the proteasome inhibitor MG132 (50 μM) for 30 min at 37 C.The proteasome activity was recorded as a fluorescent signal of amino-4-methylcoumarin release.

Real-time RT-qPCR
RT-qPCR was carried out as previously described (29).In summary, total RNA was isolated using the Roche TriPure Isolation (Manheim), and first-strand complementary DNA was generated using random primers and M-MLV reverse transcriptase (Promega).The Roche Light-Cycler 480 real-time PCR machine (Roche) was used for the RT-qPCR.Individual gene expression was normalized to 36B4 expression.Table S1 contains a list of real-time RT-qPCR primers.

Translation efficiency assay
The translation efficiency experiment was carried out as previously described (18).In brief, 200 mg liver samples from Wtap flox/flox and Wtap-HKO mice were isolated, immediately frozen, and pestled into powder in liquid nitrogen.About 600 μl polysome lysis buffer (20 mM Tris-HCl [pH 7.4], 150 mM NaCl¸5 mM MgCl2, 1% Triton-100, 1× Protease inhibitor cocktail, 1 mM DTT, 100 μg/ml cycloheximide, 1 U/ μl SUPERase In RNase Inhibitor) was added.The samples were rotated and triturated before being centrifuged at 13,000 rpm for 10 min at 4 C.The supernatant was analyzed for absorbance at 260 nm using NanoDrop 2000 before being transferred to a sucrose gradient (15%-55% sucrose gradient prepared by The Gradient Master) for ultracentrifugation at 38,000 rpm for 3 h at 4 C.The Piston Gradient Fractionator was used to examine and collect ribosome fractions.Total RNA was isolated from each fraction, and RT-PCR was performed.

ChIP assays
Liver perfusion with 1% paraformaldehyde was used to fix the livers.The nuclei were extracted from livers and sonicated (M220 Focused-ultrasonicator; Covaris) to break genomic DNA into 500-to 1000-bp fragments using a chromatin shearing kit (520127; truChIP Chromatin Shearing Kit; Covaris).Anti-RNA polymerase II antibodies (Rpb1 CTD, catalog no.: 2629; Cell Signaling Technology) were used to immunoprecipitate the samples.DNA was extracted, and qPCR analysis was performed.Table S1 contains a list of qPCR primers.

Statistical analysis
The data were reported as means ± SD.Student's t tests were used to compare the differences between two groups.The differences between three groups were examined using one-factor ANOVA, and the least significance difference t test.P <0.05 was considered statistically significant in all analyses.Kolmogorov-Smirnov test for normality with p >0.1 suggested the samples followed a normal distribution.Statistical analyses and figures were made using GraphPad Prism version 6.02 (GraphPad Software, Inc).ANOVA and least significance difference t test analyses were performed using SPSS 21.0 (SPSS, Inc).*p <0.05 and **p <0.01.

Figure 1 .Figure 2 .
Figure 1.Hepatocyte-specific Wtap depletion causes hepatocellular carcinoma (HCC) in mice.At 2 weeks old, male Wtap-HKO mice and Wtap flox/flox control mice were given a single intraperitoneal dose of DEN (50 mg/kg).The mice were fed an NC diet.Mice were sacrificed at the age of 35 weeks for HCC examination.A, the weights of the livers were measured (n = 12-16 for each group; p < 0.0001).B, representative images of mouse livers from the indicated genotype were shown.C and D, tumor number and maximal tumor size were determined (n = 9 for each group; p < 0.0001).E, representative Ki67 staining images were shown.The number of Ki67-positive cells was counted (n = 6 for each group; p < 0.0001).Data represent the mean ± SD.Significance was determined by unpaired twotailed Student's t test analysis.**p < 0.01.DEN, diethylnitrosamine; NC, normal chow diet; Wtap-HKO, hepatocyte-specific Wtap knockout.

Figure 5 .
Figure 5. Wtap deficiency in hepatocytes reduces the expression of proteasome-associated genes, which contributes to higher protein stability of GRB2 and ERK.A and B, at 8 weeks old, RNA-Seq analysis was done in the livers of Wtap flox/flox and Wtap-HKO mice (n = 3 for each group, GSE168850).Kyoto Encyclopedia of Genes and Genomes pathway analysis of the downregulated genes in Wtap-HKO mice (A).A heatmap of downregulated proteasome-related genes was displayed (B).C, proteasome activity assay was measured in the livers of Wtap flox/flox and Wtap-HKO mice (n = 6 mice per group; p < 0.0001).D-F, PSMB4 and PSMB6 protein levels in Wtap flox/flox and Wtap-HKO livers (n = 7 mice per group; PSMB4: p = 0.0233; PSMB6: p = 0.0085) and hepatocytes (n = 3 independent cell samples for each group; PSMB4: p = 0.0151; PSMB6: p = 0.0363) were determined by immunoblotting.G and H, primary hepatocytes were isolated from WT mice and treated with or without MG132.GRB2, ERK1/2, and Tubulin protein levels were determined by immunoblotting and quantified using ImageJ (n = 3 independent samples for each group; GRB2: p = 0.0262; ERK1/2: p = 0.0038).Differences between two groups were analyzed by Student's t tests.*p< 0.05 and **p< 0.01.Data represent the mean ± SD.ERK, extracellular signal-regulated kinase.