PRMT1 methylates METTL14 to modulate its oncogenic function

Abstract

N6-methyladenosine (m6A), the most abundant mRNA modification in mammalian cells, is responsible for mRNA stability and alternative splicing. The METTL3-METTL14-WTAP complex is the only methyltransferase for the m6A modification. Thus, regulation of its enzymatic activity is critical for the homeostasis of mRNA m6A levels in cells. However, relatively little is known about the upstream regulation of the METTL3-METTL14-WTAP complex, especially at the post-translational modification level. The C-terminal RGG repeats of METTL14 are critical for RNA binding. Therefore, modifications on these residues may play a regulatory role in its function. Arginine methylation is a post-translational modification catalyzed by protein arginine methyltransferases (PRMTs), among which PRMT1 preferentially methylates protein substrates with an arginine/glycine-rich motif. In addition, PRMT1 functions as a key regulator of mRNA alternative splicing, which is associated with m6A modification. To this end, we report that PRMT1 promotes the asymmetric methylation of two major arginine residues at the C-terminus of METTL14, and the reader protein SPF30 recognizes this modification. Functionally, PRMT1-mediated arginine methylation on METTL14 is likely essential for its function in catalyzing the m6A modification. Moreover, arginine methylation of METTL14 promotes cell proliferation that is antagonized by the PRMT1 inhibitor MS023. These results indicate that PRMT1 likely regulates m6A modification and promotes tumorigenesis through arginine methylation at the C-terminus of METTL14.

Introduction

The N6-methyladenosine (m6A) modification is one of the most abundant modifications of mRNA in mammalian cells [1], [2], [3]. This process is reversible and is catalyzed by a multi-component methyltransferase complex and several demethylases [4], including FTO [5,6] and ALKBH5 [7]. The m6A methyltransferase complex consists of three major proteins, namely METTL3 [8], METTL14 [8] and WTAP [9], in which METTL3 provides the catalytic core. On the other hand, the pseudo-enzyme METTL14 has no catalytic activity despite containing a methyltransferase domain [10,11]. As an RNA-binding platform, METTL14 recognizes the substrates and also activates the catalytic ability of METTL3 [12]. In particular, the non-structural C-terminus of METTL14, which is critical for mRNA binding, is also essential for m6A modification and is tightly regulated by post-translational modifications (PTMs) [13,14]. Furthermore, the aberrant expression of METTL14 is closely associated with tumorigenesis and tumor progression in multiple human cancers [12,15,16].

Arginine methylation is a relatively less studied type of PTM catalyzed by protein arginine methyltransferases (PRMTs) [17], [18], [19]. Three types of PRMTs catalyze different arginine methylation on histone and other non-histione substrates by consuming S-adenosylmethionine (SAM) [20,21]. Type I PRMTs, including PRMT1, PRMT3, PRMT4, PRMT6, and PRMT8, catalyze arginine monomethylation and asymmetric dimethylation [22,23]. Type II PRMTs, including PRMT5 and PRMT9, catalyze monomethylation and symmetric dimethylation of arginine, while the only type III PRMT, PRMT7, catalyzes monomethylation of arginine [17,20]. Arginine methylation usually occurs on the DNA- or RNA-binding proteins, such as Histone H3 [24,25], Histone H4 [24], CHTOP [26], PABP1 [27], FOXO [28], [29], [30], and RNA helicase [31], and plays an oncogenic role in human cancers [32,33]. Therefore, specific inhibitors of PRMTs have been developed for cancer treatment, including the PRMT1 inhibitor GSK3368715 and PRMT5 inhibitors GSK3326595, PF-06939999, PRT811, and JNJ-64619178 in clinical trials [34], [35], [36], [37]. Methylated arginine could be recognized by several methylation reader proteins, such as TDRD3 [38], SMN [39], and SPF30 [40], or removed by the Jmjd6 (Jumonji domain-containing 6), a histone protein arginine demethylase, although its exact role as an arginine demethylase is currently under debate [41].

The arginine/glycine-rich (RGG/RG) polypeptides are preferentially methylated by PRMT1, and many reported PRMT1 substrates harbor multiple tandem RGG repeats [17]. Notably, the non-structural C-terminus of METTL14 is arginine/glycine-rich, with seven tandem RGG repeats, in which the arginine residues provide a positive charge to interact with the negatively charged phosphate backbone of the mRNA [16,42]. Therefore, mutation or modification of these critical arginine residues could disrupt the METTL14-mRNA interaction, thus disturbing the METTL3-METTL14-WTAP complex-mediated m6A modification. Previous studies have shown that PRMT1 methylates METTL14 at its C-terminus, but the primary methylation site has not been determined yet [13,14].

To this end, we have found that the methyltransferase PRMT1, but not other PRMT family members, methylates METTL14 at its C-terminal RGG repeats, and the R442 and R445 residues are likely the major arginine methylation sites. As such, PRMT1 inhibition or mutation of the R442 and R445 residues leads to impaired METTL14 methylation, thereby disrupting the m6A modification of mRNA, which subsequently inhibits tumor cell proliferation. This regulation likely provides a rationale for using PRMT1 inhibitor as an alternative approach to target the m6A modification.

Materials and methods

Cell culture and reagents

HEK293T and HeLa cells were purchased from American Type Culture Collection (ATCC) and maintained in Dulbecco's Modified Eagle's Medium (DMEM, Gibco) containing 10% fetal bovine serum (FBS, Gibco), 100 Units of penicillin and 100 µg/mL streptomycin (Gibco). All cell lines were routinely tested for mycoplasma contamination every 2 months using the MycoAlert kit (Lonza).

Immunoblot (IB) and Immunoprecipitation (IP)

Cells were lysed in EBC Buffer (50 mM Tris PH 7.5, 120 mM NaCl, 0.5% NP40) containing protease inhibitor cocktail (A32963, ThermoFisher) and phosphatase inhibitor (K1015, APExBIO) as described before [43]. Protein concentrations were determined using the Bio-Rad protein assay dye reagent (Bio-Rad). Equal amounts of proteins were resolved by SDS-PAGE and transferred to a polyvinylidene difluoride membrane (PVDF, Bio-Rad). After being blocked with 5% fat-free milk, membranes were incubated with indicated primary antibodies at 4°C overnight. The membrane was washed four times with TBST buffer and incubated with appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies for another 1 h at room temperature before detection with HRP Substrate Luminol Regent (Millipore). For immunoprecipitation, 1.0 mg of whole cell lysates were incubated with 10 µL of primary antibody-conjugated beads for 4 h at 4°C. Nonspecific bindings were removed by washing three times with NETN buffer (20 mM Tris PH 8.0, 100 mM NaCl, 0.5% NP40, 1 mM EDTA). Bead-bound proteins were resolved by SDS-PAGE and subsequently immunoblotted with indicated antibodies. Antibodies used are listed below: anti-Flag (F2555), anti-Flag agarose beads (A2220), anti-HA (04-902), anti-HA agarose beads (A2095), anti-Vinculin (V9131), peroxidase-conjugated anti-mouse secondary antibody (A4416), and peroxidase-conjugated anti-rabbit secondary antibody (A4914) were purchased from Sigma-Aldrich. Anti-GST (2625) and anti-Asymmetric Di-Methyl Arginine (13522) were obtained from Cell Signaling Technology.

Lentiviral packaging and transduction

For viral production, lentiviral constructs were transfected into HEK293T cells together with pMD2.G and psPAX2 helper plasmids using PEI (23966, Polysciences). Forty-eight hours post-transfection, viral supernatants were collected. HeLa cells were incubated with 1 mL viral supernatant with 4 μg/mL polybrene (Sigma-Aldrich) in a final volume of 2 mL for 24 h and then supplemented with 2 mL fresh medium for an additional 48 h. Gene expression changes were assessed after selecting cells in 1 mg/mL puromycin (Thermo Fisher Scientific).

Colony and sphere formation assays

The colony formation assay was performed using the protocol described before [44]. Briefly, cells were seeded into 6-well plates in equal numbers and cultured for 2 weeks. Subsequently, cells were fixed by fixing buffer (acetic acid/methanol = 1:7) and stained with 0.4% crystal violet in 20% ethanol. The numbers of colonies were quantified by the Image J software. Sphere formation assay was conducted by plating cells into a 6-well ultra-low adherent plate (1 × 10³ cells per well) (Stemcell Technologies Inc) for culture of 3 weeks.

Cell Viability Assay

HeLa cells (3 × 10³) were seeded into 96-well microtiter plates with 100 µL medium. After adhering, cells were treated with 200 nM of PRMT1 inhibitor MS023 (Selleckchem). For the viability assay, 10 µL of Cell Counting Kit-8 (CCK-8) reagent (APExBIO) was added into each well, and cells were cultured for another 2 h at 37°C. Subsequently, the OD₄₅₀ values were measured using the multi-mode microplate reader.

Statistical analysis

Data were presented as mean ± SD and analyzed by one-way or two-way ANOVA with Tukey's correction using GraphPad Prism 7 software. Differences were considered significant when p < 0.05.

Results

PRMT1 interacts and methylates METTL14

Post-translational methylation of proteins is critical for protein stability and enzymatic activity. To test whether METTL14 can be methylated by arginine methyltransferase(s), we generated a stable HEK293T-HA-METTL14 knock-in cell line by using the CRISPR/Cas-9 approach to insert 3 tandem HA-tag repeats after the ATG code of the first methionine of METTL14 locus, which enabled to detect METTL14 at endogenous basal level (Fig. 1A). Using the stable HEK293T-HA-METTL14 cell line, we overexpressed individual GFP-tagged PRMT family members to test which PRMT interacts with METTL14. The co-immunoprecipitation assay showed that endogenous METTL14 interacted with PRMT1 and, to a lesser extent, PRMT2, but not other PRMTs we tested (Fig. 1B). More importantly, PRMT1 overexpression led to asymmetric arginine dimethylation (ADMA) of endogenous METTL14, while PRMT2 or other PRMT family members did not (Fig. 1B). Because we used a pan-ADMA antibody to detect the arginine methylation of METTL14, we reasoned that a false positive signal could be detected due to ADMA modification on a METTL14-binding protein with a similar molecule weight instead of METTL14 itself. To exclude this possibility, we generated 3 types of tagged-METTL14 constructs with a Flag, HA, or GST, respectively, whose molecular weights were different and could be easily distinguished in blotting. When ectopically overexpressed in HEK293T cells, all three tagged METTL14 proteins showed ADMA modification (Fig. 1C), validating the methylation of METTL14 by PRMT1 in this experimental setting.

Fig. 1.

PRMT1 interacts and methylates METTL14. A. A schematic diagram depicting the HA-tag knocked into the METTL14 locus in HEK293T cells. B. PRMT1 promotes the asymmetric dimethylation of METTL14. Immunoblot (IB) analysis of the whole cell lysis (WCL) and HA-immunoprecipitant (IP) derived from HA-METTL14-knock-in HEK293T cells expressing GFP-tagged PRMTs. ADMA: asymmetric dimethylarginine. C. METTL14 with different tags shows similar asymmetric dimethylation status. IB analysis of HA, Flag, or GST-tagged METTL14 derived from HEK293T cells transfected with indicated METTL14 constructs. D. Arginine methylation of METTL14 likely depends on the enzymatic activity of PRMT1. IB analysis of WCL and HA-IP derived from HA-METTL14-knock-in HEK293T cells expressing GFP-PRMT1-WT or E162 mutants. E. METTL3 promotes the stability and methylation of METTL14. IB analysis of WCL and HA-IP derived from HEK293T cells transfected with indicated constructs.

PRMT1 has a catalytic core in which the glutamine 162 (E162) residue is critical for its methyltransferase activity, and mutants such as E162Q and E162A that perturb PRMT1 activity are usually used as catalytic-dead mutants [45]. We found that only the wild-type (WT) PRMT1, but not the catalytic-dead mutants PRMT1-E162Q and E162A, methylated METTL14 (Fig. 1D), indicating that PRMT1 promotes the asymmetric arginine methylation of METTL14 in an enzymatic activity-dependent manner.

As a core component of the m6A catalyzing complex, METTL3 binds with METTL14 and protects METTL14 from STUB1-mediated degradation [46]. To determine whether the presence of METTL3 regulates the methylation of METTL14, we co-expressed Flag-METTL3, HA-METTL14, and GFP-PRMT1, and found that PRMT1 promoted the asymmetric dimethylation of METTL14 when co-expressed with METTL3 (Fig. 1E). These results support the notion that METTL14 is likely a bona fide substrate of PRMT1.

PRMT1 methylates METTL14 on its C-terminal RGG repeats

To further identify the PRMT1-catalyzed arginine methylation site(s) of METTL14, we aligned the METTL14 protein sequence among various species. Interestingly, we identified the seven conserved RGG domains in the C-terminus of METTL14, the potential arginine methylation sites matching the PRMT1 consensus motif (Fig. 2A). To determine whether PRMT1 methylates METTL14 at these RGG repeats, we constructed two different METTL14 truncates by either deleting the last 4 RGG repeats (hereafter named METTL14 (1-434)) or all the 7 RGG repeats (hereafter named METTL14 (1-402)) (Fig. 2B). Next, we performed an in-cell methylation assay using HA-METTL14 WT or indicated deletion mutants together with GFP-PRMT1 in HEK293T cells. We found that METTL14 (1-434) largely reduced PRMT1-mediated arginine asymmetrically methylation, while METTL14 (1-402) completely abolished the residual methylation band, indicating that all these RGG repeats could be potentially methylated by PRMT1, in this experimental setting (Fig. 2C). Furthermore, PRMT1 remained to interact with METTL14 regardless of deleting the C-terminal region (402-456), suggesting that PRMT1 interaction site in METTL14 likely differs from the methylated residues targeted by PRMT1 (Fig. 2C). These results indicate that all these RGG repeats could be potentially methylated by PRMT1 (Fig. 2C), although there may be other potential methylated residues in this C-terminal region because the antibody may not recognize all the methylated arginines.

Fig. 2.

PRMT1 binds the C-terminus of METTL14. A. Sequence alignment of the putative arginine/glycine-rich motif in METTL14. B. A schematic illustration of METTL14 mutants with RGG repeats deletion. C. PRMT1 methylates arginine residue(s) in RGG repeats of METTL14 in cells. IB analysis of WCL and HA-IP derived from HEK293T cells expressing GFP-PRMT1 together with HA-METTL14-WT, HA-METTL14 (1-402), or HA-METTL14 (1-434).

PRMT1 mainly asymmetrically methylates the R442 and R445 residues of METTL14

Having demonstrated that PRMT1 methylates its substrate at the conserved RGG/RG motif, and the sequential C-terminal deletion revealed that the last 4 RGG repeats of METTL14 are the potential methylation sites, we next sought to determine the major arginine asymmetric dimethylation residues of METTL14 by PRMT1. To this end, we used the GPS-MSP tool (http://gps.biocuckoo.cn/) to predict the potential methylation sites [47]. Although most C-terminal RGG repeats are predicted to be possible modification sites (Fig. 3A), our results suggest that deleting the last four RGG repeats largely abolished the methylation event. Thus we focused on the arginine residues in the last four RGG repeats (Fig. 2). To this end, we generated mutants to replace single arginine residue with lysine of the last four RGG repeats, namely R438K, R442K, R445K and R450K, and the 4RK (R438K/R442K/R445K/R450K) that all the four arginines were replaced (Fig. 3B). Co-expression of PRMT1 promoted the ADMA modification of METTL14 WT, R438K, and R450K mutants, while R442K, R445K, and 4RK mutants almost completely abolished PRMT1-mediated METTL14 asymmetric dimethylation (Fig. 3C). Taken together, these data suggest that PRMT1 asymmetrically dimethylates the 442 and 445 arginine residues of METTL14 (Fig. 3D).

Fig. 3.

PRMT1 methylates the arginine 442 and 445 residues of METTL14. A. Predicted arginine methylation sites of human METTL14 protein. The prediction is conducted by GSP-MBP. B. A schematic diagram illustrates the mutation of putative methylated arginine residues in METTL14. C. PRMT1 methylates METTL14 on R442 and R445 residues. IB analysis of WCL and HA-IP derived from HEK293T cells expressing GFP-PRMT1 together with HA-METTL14 or the indicated R-to-K mutants. D. A schematic diagram illustrates the two major methylated arginine residues of METTL14.

Methylated METTL14 is recognized by SPF30

As we found PRMT1 as a METTL14 methylation writer, we further sought to identify the eraser and reader of METTL14 methylation. Arginine-methylated proteins are recognized by the arginine methylation readers, such as TDRD3, SMN1 and SPF30 [40,48,49]. To determine which methyl-arginine reader protein recognizes methylated METTL14, we co-expressed METTL14 together with TDRD3, SMN1, or SPF30, in HEK293T cells (Fig. 4A). We found that methylated METTL14 majorly bound with SPF30, and TDRD3 to a lesser extent, but not SMN1, in cells (Fig. 4A). To further identify whether the interaction between METTL14 and SPF30 depends on the arginine methylation of the C-terminal RGG repeats in METTL14, we overexpressed SPF30 together with METTL14-WT or different mutants and found that SPF30 bound with METTL14-WT, but not METTL14 (1-402) lacking all the RGG repeats (Fig. 4B). On the other hand, the reduction in METTL14/SPF30 interaction was observed when the METTL14 (1-434) mutant that lacks the last 4 RGG was expressed (Fig. 4B). Furthermore, with the increased expression of Jmjd6 demethylase, the PRMT1-mediated asymmetric dimethylation of METTL14 was gradually decreased (Fig. 4C), implying that Jmjd6 might be the methylation eraser of METTL14. However, given that the arginine demethylase role of Jmjd6 is currently debated, additional in-depth investigation is warranted to reveal whether Jmjd6 demethylates METTL14 in cells.

Fig. 4.

SPF30 and Jmjd6 are likely the reader and eraser of METTL14 methylation. A. Methylated METTL14 is recognized by SPF30. IB analysis of WCL and HA-IP derived from HEK293T cells expressing HA-METTL14 together with Flag-tagged TDRD3, SMN1, or SPF30. B. RGG repeats are required for METTL14 to bind with SPF30. IB analysis of WCL and HA-IP derived from HEK293T cells expressing Flag-SPF30 together with HA-METTL14-WT, HA-METTL14 (1-402), or HA-METTL14 (1-434). SE: Short exposure, LE: Long exposure. C. The arginine methylation of METTL14 is likely removed by Jmjd6. IB analysis of WCL and HA-IP derived from HEK293T cells expressing HA-METTL14 together with GFP-PRMT1 and Flag-Jmjd6.

Arginine methylation on the C-terminus of METTL14 promotes cancer cell proliferation

Having identified the methylation writer, eraser, and reader of METTL14, we next sought to analyze the potential biological role of arginine methylation of METTL14. To this end, we generated stable HeLa cell lines ectopically expressing either METTL14-WT, METTL14 (1-434), or METTL14 (1-402). We found that ectopic expression of METTL14-WT promoted tumor cell growth compared with control, while the RGG deletion mutants inhibited cell growth (Fig. 5A). Similarly, the METTL14 (1-434) and METTL14 (1-402) mutants reduced HeLa cell viability compared to METTL14-WT (Fig. 5B). Additionally, METTL14-WT promoted sphere formation compared to the control or RGG deletion mutants (Fig. 5C–E). In keeping with the results of the cell viability and proliferation, the colony formation assay demonstrated that METTL14-WT promoted the colony formation of HeLa cells, while METTL14 (1-434) and METTL14 (1-402) mutants suppressed it in this experimental setting (Fig. 5F, G). These results indicate that the C-terminal RGG motif is likely required for the oncogenic role of METTL14.

Fig. 5.

The methylation on the C-terminus of METTL14 promotes cell proliferation and transformation ability. A. METTL14 promotes cell proliferation, and deletion of its C-terminal RGG repeats compromises its oncogenic role. HeLa cells stably expressing METTL14-WT, METTL14 (1-434), or METTL14 (1-402) mutants were plated in a 6-well plate and subjected to cell growth curve assay. B. Deletion of the C-terminal RGG repeats compromises the role of METTL14 in promoting cell proliferation. HeLa cells, as in A, were subjected to cell viability assay. C-E. METTL14 facilitates sphere formation, and deletion of the C-terminal RGG repeats compromises the role of METTL14 in promoting the sphere formation. Representative sphere images are shown. Scale bar, 100 μm (C). Sphere size was measured by ImageJ (D). Spheres were counted under a microscope (E). F. METTL14 RGG repeats facilitate colony formation of HeLa cells. G. Deletion of the C-terminal RGG repeats compromise the role of METTL14 in promoting the colony formation of HeLa cells. ANOVA with Tukey's correction. *p < 0.05, **p < 0.01. (A, B, D, E, and G).

PRMT1 inhibitor treatment largely abolishes endogenous METTL14 methylation and suppresses METTL14-mediated cancer cell proliferation

Given that the C-terminus of METTL14 is necessary for RNA binding, the decreased cell viability by RGG repeats deletion might depend on arginine methylation. Therefore, we attempted to determine whether the observed oncogenic function of METTL14 is related to PRMT1-mediated arginine methylation. To this end, we generated HeLa HA-METTL14 knock-in cells in which 3 × HA is inserted after the first methionine of METTL14 to analyze endogenous METTL14 methylation. We treated resulting cells expressing HA-tagged endogenous METTL14 with the PRMT1 inhibitor MS023 [50] and conducted an in-cell methylation assay. We found that endogenous METTL14 was subjected to arginine asymmetric dimethylation and further demonstrated that the methylation was abolished by PRMT1 inhibitor treatment (Fig. 6A). Next, to test the effects of PRMT1 inhibitor treatments on METTL14-mediated cell proliferation, we treated HeLa cells stably expressing METTL14-WT or indicated truncation mutants with the PRMT1 inhibitor. Notably, we found that PRMT1 inhibitor significantly inhibited the proliferation of HeLa cells expressing METTL14-WT, but not METTL14 (1-434) or METTL14 (1-402) mutant (Fig. 6B). Consistent with the cell proliferation results, we further showed that PRMT1 inhibition inhibited colony formation of HeLa cells ectopically expressing METTL14-WT, while expressing the RGG deletion mutants showed relatively little effects in response to PRMT1 inhibitor treatment (Fig. 6C, D). Taken together, these results demonstrate that the PRMT1-mediated arginine methylation on the C-terminal RGG repeats of METTL14 plays a potential role in cancer cell proliferation. However, further in-depth studies are required to explore whether PRMT1-mediated METTLE14 methylation at its C-terminus affects the oncogenic role of METTL14 in vivo, which might provide the molecular basis or rationale for using PRMT1 inhibitor as an efficient anti-cancer treatment.

Fig. 6.

PRMT1 inhibitor treatment largely abolishes METTL14 methylation and suppresses METTL14-mediated cancer cell proliferation. A. Endogenous METTL14 is subjected to arginine asymmetric dimethylation, which is reversed by PRMT1 inhibitor treatment. IB analysis of WCL and HA-IP derived from HeLa cells with 3 × HA knock-in at METTL14 locus treated with or without the PRMT1 inhibitor MS023 (200 nM). B. PRMT1 inhibition compromises the cell viability of HeLa cells expressing METTL14-WT, but not cells expressing METTL14 (1-434) or METTL14 (1-402) mutant. HeLa cells stably expressing METTL14-WT, METTL14 (1-434), or METTL14 (1-402) mutant were treated with DMSO or PRMT1 inhibitor MS023, and the cell number were counted at the indicated time. C-D. PRMT1 inhibition compromises the colony formation of HeLa cells expressing METTL14-WT, but not METTL14 (1-434) or METTL14 (1-402) mutant. HeLa cells, as in B, were treated with DMSO or PRMT1 inhibitor MS023 and then subjected to colony formation assay (C). The data shown represent the means ±SD of biological triplicates (D). ANOVA with Tukey's correction. *p < 0.05, **p < 0.01. (B and D).

Discussion

PRMT1 catalyzes the asymmetric arginine dimethylation of protein substrates to promote tumorigenesis, and its substrates have an RGG/RG motif, such as FUS and Nucleolin [51,52]. In the human genome, there are only a few proteins with such tandem RGG motifs, and METTL14 is one of them. On the other hand, PRMT1 substrates are usually DNA/RNA-binding proteins, and METTL14 is the key subunit of the m6A methylation complex for RNA binding, suggesting METTL14 as a potential PRMT1 substrate.

Previous studies showed that METTL14 is methylated by PRMT1 and PRMT6 in vitro but only by PRMT1 in cells, and mutation of all the 13 arginine residues in the C-terminus abolished the methylation of METTL14, compromising its function in m6A production in mouse embryonic stem cells (mESC) [14]. Moreover, arginine methylation of METTL14 is essential for its interaction with RNAPII and translation of DNA repair genes, and thus, disruption of METTL14 methylation sensitizes cells to DNA damage [14]. Another independent study showed that PRMT1 methylates METTL14 at the R255 to promote the association between WTAP and the METTL3/METTL14 subcomplex, thereby promoting the activity to produce m6A modification in mESCs, which is essential for the endoderm differentiation of normal mESCs [13]. However, little is known about whether and how PRMT1-mediated METTL14 methylation regulates tumorigenic function.

In this study, we found that PRMT1 methylates METTL14 at its C-terminal RGG repeats, and R442 and R445 are two major arginine methylation sites. Furthermore, we found that methylated METTL14 can be recognized by the arginine methylation reader protein SPF30, and the arginine methylation is likely removed by Jmjd6. METTL14 functions as an oncoprotein associated with a variety of cancers, and arginine methylation at its C-terminus might be critical for its oncogenic function [12]. A previous study demonstrated that an increased m6A modification occurs in more than half of pancreatic cancer [53], and upregulation of METTL14 mediates the elevation of PERP mRNA m6A modification, which promotes the growth and metastasis of pancreatic cancer [53]. In pancreatic cancer, inhibition of METTL14 also enhances the sensitivity of tumor cells to cisplatin via the mTOR signaling pathway and subsequently induces apoptosis and autophagy [54]. Additionally, by regulating the localization of subcellular proteins, METTL14 overexpression elevates m6A levels and enhances the progression of prostate cancer [55]. METTL14-mediated m6A modification can also enhance the oncogenic LNCAROD lncRNA in head and neck squamous cell carcinoma (HNSCC) cells and contribute to HNSCC oncogenesis and migration [56]. These studies demonstrate the importance of METTL3/METTL14 inhibition in cancer treatment. However, it should be noted that the role of the METTL3/METTL14 complex in tumorigenesis appears to be tissue context and cell type dependent [12,57]. Since PRMT1 is a well-defined oncoprotein, and its inhibitor has already been advanced into clinical trials [58], [59], [60], PRMT1 inhibitor could be used as an effective alternative approach for m6A-targeted therapy. Recent studies have shown that PRMT1 regulates cancer immunity by modulating RNA alternative splicing [61]. Thus, its role through methylation of METTL14 could potentially play a key role in augmenting clinical benefit, which awaits further in-depth investigation.

Data availability

All data are contained within the article.

CRediT authorship contribution statement

Jingchao Wang: Methodology, Data curation, Writing – original draft, Writing – review & editing. Zhen Wang: Methodology, Writing – review & editing. Hiroyuki Inuzuka: Methodology, Writing – review & editing. Wenyi Wei: Methodology, Data curation, Writing – review & editing. Jing Liu: Methodology, Data curation, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.