Recent advances in small molecules targeting the METTL3

ABSTRACT

N⁶-methyladenosine (m⁶A) is the most prevalent internal modification of eukaryotic mRNA, playing a crucial role in the regulation of gene expression. Methyltransferase-like 3 (METTL3), a key catalytic component of the m⁶A methyltransferase complex, is primarily responsible for the deposition of m⁶A on target RNA. Recent studies have revealed that METTL3 contributes to diverse pathological processes, particularly tumorigenesis, through both m⁶A-dependent and independent mechanisms. As a result, METTL3 has attracted increasing interest as a potential therapeutic target across various cancer types. This review summarizes recent advances in the discovery of small molecules targeting METTL3, including substrate-competitive inhibitors, allosteric inhibitors, and proteolysis-targeting chimeras (PROTACs). It also discusses the strategies in their discovery, the associated structural features, and the remaining challenges and future directions in this field. Overall, these efforts provide valuable insights into the design and discovery of METTL3-targeted therapeutics with potential clinical applications.

GRAPHICAL ABSTRACT

1. Introduction

N⁶-methyladenosine (m⁶A) is the most abundant internal modifications found in eukaryolic mRNA [1], playing a pivotal role in regulating various physiological and pathological processes [2–5]. This epitranscriptomic modification is catalyzed by a multi-subunit methyltransferase complex (MTC) [6], in which Methyltransferase-like 3 (METTL3) and Methyltransferase-like 14 (METTL14) serve as the core subunits [7]. Additional cofactors, such as Wilms’ tumor 1 associated protein (WTAP) are required for proper catalysis and RNA substrate recognition [8–10]. Notably, METTL3 functions as the sole catalytic subunit, and has been implicated as a key regulator in human physiology and disease pathogenesis [11–15], particularly in tumor initiation and progression [16–18].

Accumulating evidence has demonstrated that METTL3 is frequently overexpressed in a variety of cancers, where it often acts as a tumor promoter [19–23]. It regulates a broad array of cancer-related processes, including cell proliferation, apoptosis, invasion and metastasis, primarily through m⁶A-mediated modulation of oncogenic transcripts [24,25]. For instance, genetic depletion of METTL3 in acute myeloid leukemia (AML) cell lines has been shown to result in impaired proliferation, cell cycle arrest, and enhanced apoptosis [19,22]. In bladder cancer, METTL3 facilitates tumor progression by activating the AFF4/NF-κB/MYC signaling cascade in an m⁶A-dependent manner [26]. In colorectal cancer (CRC), METTL3 maintains the expression of SRY-box transcription factor 2 (SOX2) via the m⁶A-IGF2BP2 axis and enhances metastatic potential by regulating the miR-1246/insulin-like growth factor 2 mRNA-binding protein 2 (SPRED2) pathway [21,27].

Beyond its direct effects on tumor cells, METTL3 also influences antitumor immunity. For example, Yu et al. demonstrated that suppression of METTL3 activity can remodel the tumor microenvironment into a more immunologically active state, thereby enhancing the efficacy of immune checkpoint blockage therapy with anti-PD-1 antibodies in mouse models of non-small cell lung cancer [28]. Guirguis et al. further found that inhibiting METTL3’s catalytic function leads to the accumulation of double-stranded RNA, triggering a robust intrinsic interferon response that promotes antitumor immunity [29]. In CRC, Chen et al. reported that METTL3 impairs antitumor immune responses via the m⁶A-modified basic helix-loop-helix family member e41 (BHLHE41)-CXCL1/CXCR2 signaling axis, ultimately promoting tumor progression [30].

Recent studies also indicate that METTL3 can promote cancer progression independently of its methyltransferase activity. These non-canonical roles often involve altered subcellular localization or protein-protein interactions. For instance, Gregory et al. demonstrated that cytoplasmic METTL3 can act as a reader of m⁶A marks to promote the translation of oncogenic mRNAs in lung cancer cells [31]. Similarly, Yu et al. reported that cytoplasmic METTL3 interacts with polyadenylate-binding protein 1 (PABPC1) and stabilizes its interaction with the eukaryotic initiation factor 4F (eIF4F) cap-binding complex, thereby enhancing the translation of non-m⁶A modified epigenetic factors and contributing to gastric cancer development [32]. Taken together, these findings underscore METTL3’s multifaceted role in tumor biology and its promise as a therapeutic target. While several prior reviews have discussed METTL3’s biological functions and the early development of inhibitors [33–36], this review focuses on recent progress in small-molecule modulators of METTL3, including PROTACs. We highlight targeting strategies, structure features, and biological activities, and discuss the current challenges and future opportunities in targeting METTL3 for cancer therapy. These insights aim to support the rational design and development of METTL3-targeted therapeutics for clinical applications.

2. Structure of METTL3-METTL14 complex

The METTL3-METTL14 complex, also known as the m⁶A-METTL complex (MAC), is a stable and asymmetric heterodimer with a 1:1 stoichiometric composition [7]. Both METTL3 (residues 358–580) and METTL14 (residues 165–378) contain methyltransferase domains (MTDs) (Figure 1). In addition to the MTD, METTL14 harbors two functionally distinct motifs: an N-terminal α-helical motif (NHM, residues 116–163), which bridges the heterodimer interface, and a C-terminal RGG motif (CTM, residues 380–402). Despite their structural similarity, only METTL3 possesses catalytic activity, as it binds the methyl donor S-adenosylmethionine (SAM) via a conserved Asp-Pro-Pro-Trp (DPPW) motif [37]. In contrast, METTL14 lacks SAM-binding ability and serves as a structural scaffold that supports METTL3-RNA interactions. Together, METTL3 and METTL14 exert a synergistic effect on catalytic efficiency, as each alone exhibits only minimal methyltransferase activity [38].

Figure 1.

Schematic domain and overall structure of the METTL3-METTL14 heterodimer (PDB:5K7W). Color-coded bars indicate the crystallization constructs, with corresponding residue numbers labeled. The surface representation depicts METTL3 in magenta and METTL14 in cyan. S-adenosylmethionine (SAM) are shown in green stick.

The co-crystal structure of METTL3-SAM complex revealed that SAM binds deeply within the METTL3 catalytic pocket, with its purine ring establishing hydrogen bonds with the carboxylate group of Asp377 and the backbone carbonyl of Ile378 (Figure 2(A)) [39]. The ribose hydroxyl groups further reinforce SAM binding via hydrogen bonding interactions with Arg536, Asn549, and Gln550. Meanwhile, the methionine moiety coordinates a network of polar contacts with Asp395, Lys513, His538, and Asn539, collectively contributing to the stabilization of the METTL3-SAM complex. Of note, SAM binding induces a conformational rearrangement of two flexible loop regions in METTL3—Gate loop 1 (residues 396–410) and Gate loop 2 (residues 507–515) – which undergo ligand-dependent closure to enclose the methyl donor (Figure 2(B)). The degree of loop closure varies depending on the nature of the bound molecule. Small-molecule METTL3 inhibitors that target the SAM-binding pocket often feature an adenine-mimetic core to anchor within the purine-binding region, along with a hydrophobic group that can alter the gate loop conformation, allowing deeper insertion into an adjacent hydrophobic cavity. These molecular interactions between METTL3 and such inhibitors will be elaborated in the following section.

Figure 2.

Structural characterization of METTL3. (A) Crystal structure of METTL3 in complex with S-adenosylmethionine (SAM) (PDB: 5IL1). (B) Structural overlap of the ligand-free (orange loops; gray cartoon, PDB: 5IL0) and SAM-bound (green loops; gray cartoon, PDB: 5IL1) SAM binding induces a conformational rearrangement of gate loop 1 and loop 2.

3. Recent studies on small-molecule modulators targeting METTL3

Given the critical role of METTL3 in tumor progression, increasing research efforts have been directed toward the development of small-molecule modulators targeting METTL3. These compounds can be classified into three main categories based on their binding sites and modes of action: substrate-competitive inhibitors, allosteric inhibitors, and proteolysis-targeting chimeras (PROTACs).

3.1. Substrate-competitive inhibitor

The most extensively studied class of METTL3 modulators comprises molecules that competitively occupy the SAM-binding site. These compounds can be further classified according to the core scaffolds they employ to mimic SAM, particularly its adenosine moiety. In the following section, we describe this class of inhibitors, emphasizing their core scaffold (highlighted in red), which mimics the adenine moiety of SAM and facilitates key interactions with Asp377 or Ile378.

3.1.1. Adenosine scaffold

SAM analogs represent the earliest reported class of substrate-competitive inhibitors targeting METTL3. In 2020, the Caflisch group conducted a virtual screening of a commercially available compound library comprising approximately 4,000 adenosine derivatives [40]. This effort yielded 70 candidates for further biochemical evaluation, among which compound 1 exhibited moderate inhibition of METTL3 with an IC₅₀ of 8.7 μM (Figure 3(A)). Structural analysis (PDB:6TTP) revealed that compound 1 occupies the SAM-binding pocket of METTL3, where its adenine moiety forms hydrogen bonds with residues Ile378 and Asn549 (Figure 3(B)). Notably, electron density beyond the amide region was unresolved, indicating a flexible, disordered tail. In the same year, Wang and coworkers performed virtual screening of over 500,000 compounds from the ChemDiv and MCE libraries [41], ultimately identifying compound 2 (Cpd-564), which demonstrated binding affinity to METTL3 in a cellular thermal shift assay (CETSA). In parallel, Accent Therapeutics reported two classes of selective adenosine-based METTL3 inhibitors. Representative compounds 3 and 4 exhibited potent inhibitory activity with nanomolar potency (IC₅₀ < 10 nM) [42].

Figure 3.

SAM-derived METTL3 inhibitors. (A) Chemical structures of compounds 1–4. (B) Structural overlap of the complexes of METTL3 with S-adenosylmethionine (SAM) (blue, PDB code 5IL1) and compound 1 (green, 6TTP).

Despite their potential inhibitory activity against METTL3, these adenosine analogs limited by poor membrane permeability due to their hydrophilicity and frequently display limited selectivity against other SAM-dependent enzymes, thereby raising the risk of off-target effects. Together, these limitations significantly restrict their broader applicability.

3.1.2. Aminopyrimidine scaffold

To overcome the limitations associated with adenosine analogs, research efforts have shifted toward the development of non-adenosine derivatives as alternative METTL3 inhibitors. For example, following the screening of an adenosine-like compound library, the Caflisch team optimized hit compounds and ultimately identified compound 5 (UZH1a, PDB code: 7ACD) [43], a potent METTL3 inhibitor featuring an aminopyrimidine scaffold, with an IC₅₀ of 280 nM – approximately 100 times more active than its S-enantiomer, 6 (UZH1b) (IC₅₀ = 28 μM) (Supplementary Figure S1A). Moreover, UZH1a demonstrated high selectivity over other methyltransferases, retaining over 75% enzymatic activity at a 10 μM concentration, and showed favorable permeability in a Caco-2 assay (Papp > 10⁻⁵ cm/s).

Crystal structure analysis revealed that UZH1a occupies the SAM-binding pocket, with the aminopyrimidine moiety forming three hydrogen bonds with Ile378 and Asn549, and Asp377 (Supplementary Figure S1B). Additionally, its hydroxyl group forms a hydrogen bond with the side chain oxygen of Asn549 and participates in water-mediated polar interactions with Gln550 and Gly535. UZH1a binding induces a displacement of the Lys513 side chain by ~ 6 Å, enabling the piperidine nitrogen to mimic Lys513’s interaction with Asp395 and to insert into a nearby hydrophobic aromatic region (Supplementary Figure S1C). This structural rearrangement likely contributes to its high selectivity over other methyltransferases.

Subsequently, the Caflisch team pursued further structural optimizations of UZH1a [33]. Initially, they removed the phenyl and phenolic hydroxyl group, yielding compound 7 (IC₅₀ of 7 μM) (Figure 4(A)). Next, they repositioned the methylene linker on the piperidine ring from the meta- to the para-position to eliminate the chiral center and further simplified the linker by removing the carbonyl group, resulting in compound 8 (IC₅₀ = 5 μM). Then, a cyclization strategy was employed to rigidify the central scaffold by introducing a series of spirocyclic structures. These efforts led to the identification of compound 9 (PDB code: 7O0L), which forms a hydrogen bond with Gln550 through its lactam structure (Figure 4(B)) and exhibits significantly enhanced potency (IC₅₀ = 0.037 μM). Further structure-activity relationship (SAR) refinements on the benzyl and pyridine moieties resulted in the development of 10 (UZH2), a potent and selective METTL3 inhibitor with an IC₅₀ of 5 nM. UZH2 demonstrated effective cellular target engagement and reduced the m⁶A/A ratio in polyadenylated RNA in MOLM-13 and PC-3 cell lines, with EC₅₀ values of 0.7 μM and 2.5 μM, respectively. The lower cellular activity relative to biochemical potency is likely attributable to intracellular SAM competition.

Figure 4.

The discovery of compound 10 (UZH2). (A) Development of compound 10 starting from 7. (B) Crystal structure of METTL3 in complex with 9 (PDB: 7O0L).

In 2023, the Caflisch group reported a new class of aminopyrimidine derivatives through high-throughput screening [44]. Initial hits, such as compound 11 (IC₅₀ = 430 μM), were optimized by modifying substituents on the pyrimidine ring and altering the hydrophobic fragment, yielding compound 12 (IC₅₀ = 0.13 μM) (Supplementary Figure S2). Crystal structure analysis (PDB ID: 7NI9) revealed that the isopropyl group of 12 displaces three structural water molecules in the hydrophobic pocket, thereby enhancing hydrophobic complementarity. Further optimization focused on the linker connecting the hydrophobic fragment. A ring fusion strategy was employed to constrain the ligand’s bound conformation, ultimately culminating in compound 13, which incorporated an aromatic naphthylamine group and achieved nanomolar potency (IC₅₀ = 54 nM). In MOLM-13 cells, compound 13 showed weak antiproliferative activity with an GI₅₀ of 6 μM), approximately 100-fold weaker than its biochemical IC₅₀. Similar to UZH2, this discrepancy may arise from competition with high concentration of intracellular SAM [45]. Additionally, thermal shift assays confirmed that compound 13 selectively binds METTL3 over related methyltransferases METTL1 and METTL16.

3.1.3. Pyrimidinone scaffold

The pyrimidinone core represents another widely utilized scaffold among non-adenosine METTL3 inhibitors. In 2021, Storm Therapeutics reported hit compound 14 (STM1760) with an IC₅₀ of 51.7 μM through high-throughput screening of approximately 250,000 drug-like molecules (Figure 5(A)) [46]. Subsequent medicinal chemistry optimization efforts led to the development of the pyrimidinone-based compound 15 (STM2457, PDB code: 7O2I), which exhibited significantly enhanced potency, with an IC₅₀ of 16.9 nM in METTL3 enzymatic assays and a binding affinity (K_d) of 1.4 nM as measured by surface plasmon resonance (SPR). Cellular thermal-shift assays confirmed its inhibitory potency against both human (IC₅₀ = 1.52 μM) and mouse (IC₅₀ = 2.81 μM) METTL3. Furthermore, STM2457 displayed remarkable selectivity, showing negligible activity against a panel of 45 other methyltransferases and 468 kinases.

Figure 5.

The discovery of compound 15 (STM2457). (A) Development of compound STM2457. (B) Crystal structure of METTL3-METTL14 in complex with STM2457 (PDB:702I).

Structural analysis revealed that the pyrimidinone scaffold in STM2457 forms a hydrogen bond with Ile378, while the carbonyl oxygen of its amide linkage interacts with Asn549 (Figure 5(B)). A conformational shift in residue Lys513 was also observed upon binding. The antileukemic potential of STM2457 was evaluated across a panel of AML cell lines, where it significantly inhibited cell proliferation and colony formation, while sparing normal cells. In vivo studies demonstrated robust tumor suppression and prolonged survival across multiple AML mouse models. As such, STM2457 has since been widely used as a tool molecule to investigate METTL3 biology [28,30,47,48].

In 2022, STORM Therapeutics further advanced 16 (STC-15), the first METTL3 inhibitor to enter clinical trials (Table 1). Interim results from a phase I clinical trial (NCT05584111) presented at the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting demonstrated that STC-15 was well-tolerated at all tested doses [49]. Among 27 evaluable patients, the study reported an overall response rate (ORR) of 11% (three partial responses) and a disease control rate (DCR) of 63% (14 stable disease and three partial responses). Pharmacodynamic assessments revealed significant reductions in methylated polyA-RNA levels in blood samples across all cohorts, indicating effective METTL3 target engagement. Additionally, RNA transcript analyses showed activation of interferon signaling pathways and innate immune responses in treated patients. Building on these encouraging results, STORM Therapeutics plans to initiate combination studies of STC-15 with immune checkpoint inhibitors and expand its clinical applications.

Table 1.

Chemical structures of compounds 16-18.

Compound	Structures
16 (STC-15)
17
18

Beyond STORM Therapeutics, several other research institutions have developed pyrimidinone-based METTL3 inhibitors (Table 1). For instance, Sichuan Haisco Pharmaceutical Co. Ltd. disclosed a series of small-molecule METTL3 inhibitors in a patent, among which compound 17 demonstrated antiproliferative activity against MOLM-13 cells (IC₅₀ = 0.56 μM) in CellTiter-Glo assays and favorable oral bioavailability (92.7%) in murine models [50]. Importantly, it exhibited low risk for cardiac toxicity, with minimal inhibition of the human ether-a-go-go-related gene [51] (IC₅₀ > 30 μM). Additionally, researchers from the First Affiliated Hospital of Zhengzhou University disclosed compound 18, which exhibited potent METTL3 inhibitory activity (IC₅₀: 0.1–100 nM) and antiproliferative effects against MOLM-13 and SK-OV-3 cell lines, with IC₅₀ values ranging from 1–20 μM [52].

3.1.4. Indazole scaffold

In addition to their previously reported pyrimidinone-based inhibitors, STORM Therapeutics has also developed a distinct series of METTL3 inhibitors based on an indazole scaffold. Among them, compound 19 (STM3006, PDB code: 8BN8) demonstrated significantly improved biochemical and cellular potency compared to STM2457 [29]. Biochemical assays showed that STM3006 inhibited METTL3 with an IC₅₀ of 5 nM, approaching the lower detection limit of the assay. SPR analysis further confirmed its high binding affinity, yielding a dissociation constant (K_d) of 55 pM (Supplementary Figure S3A).

X-ray crystallographic analysis revealed that the indazole core of STM3006 forms two hydrogen bonds with Ile378 and Asp377—the latter interaction being absent in the STM2457-bound complex structure. Additionally, its 4,4-dimethylpiperidyl group occupies a hydrophobic pocket, while the piperidine nitrogen forms a salt bridge with Asp395, collectively contributing to its improved binding affinity (Supplementary Figure S3B).

Functionally, STM3006 reduced m⁶A modification levels on polyA-RNA with an IC₅₀ of 25 nM, representing a 20-fold improvement over STM2457. It also demonstrated potent antiproliferative activity across multiple cancer cell lines. Importantly, it displayed remarkable selectivity, exhibiting selectivity over 1,000-fold specificity for METTL3 relative to a panel of 45 other RNA, DNA, and protein methyltransferases.

3.1.5. Naphthyridinone scaffold

858 Therapeutics, Inc. reported the first METTL3 inhibitor featuring a naphthyridinone scaffold [53]. Representative compound 20 demonstrated potent inhibition of METTL3-METTL14 enzymatic activity in TR-FRET assays (IC₅₀ < 100 nM) and exhibited antiproliferative effects against MOLM-13 cells (IC₅₀ < 1 μM) (Supplementary Figure S4A). Our molecular docking studies revealed that the naphthyridinone core of compound 20 forms two hydrogen bonds with residues Ile378 and Asn549, while its 6-oxa-2-azaspiro[3.4]octane moiety extends toward the solvent-exposed region (Supplementary Figure S4B).

Subsequently, STORM Therapeutics also developed a series of compounds utilizing this naphthyridinone scaffold [54]. Among these, compound 21, which shares the same aromatic linker and hydrophobic side chain found in the clinical candidate STC-15, inhibited METTL3-METTL14 activity with an IC₅₀ of 10.3 nM. In cellular assays, compound 21 displayed antiproliferative activity in Kasumi-1 cells (IC₅₀ = 947 nM) and Caov3 cells (IC₅₀ = 312 nM).

Epics Therapeutics SA recently reported the discovery and optimization of a novel series of METTL3 inhibitors featuring pyridine scaffolds (Supplementary Figure S5) [55]. Utilizing a scaffold-hopping strategy guided by QSAR modeling, they identified compound 22, an indazole derivative that exhibited potent inhibition of the METTL3-METTL14 in the scintillation proximity assay (SPA) (IC₅₀ = 2 nM) and demonstrated antiproliferative activity across multiple cancer cell lines. To improve the physicochemical properties, the indole ring in compound 22 was replaced with a smaller pyridine moiety, yielding compound 23. This modification reduced lipophilicity while maintaining lipophilic ligand efficiency (LLE), albeit with a slight decrease in potency. Subsequent SAR studies focused on modification on the aromatic linker. Among the resulting derivatives, compound 24, which incorporates a pyridone linker, demonstrated potent antiproliferative activity in Kasumi-1 cells and favorable LLE, along with a clearance rate of 854 mL/min/kg.

Further optimization involved α-methylation at the benzylic position, leading compound 25. This derivative retained strong protein interactions and exhibited improved free drug fraction and intravenous pharmacokinetic profiles, as evidenced by lower unbound clearance and higher unbound AUC in rats compared to compound 24. Additional efforts concentrated on modifying the pyridine moiety. For example, replacing the methoxy group with an electron-donating N, N-dimethylamino group at the meta-position of the pyridine ring generated compound 26, which showed over a tenfold improvement in antiproliferative activity against Kasumi-1 cells compared to 25. This modification highlights the critical role of substitutions on the pyridine scaffold in improving cellular potency. Fine-tuning around compound 26 culminated in the identification of compound 27 (EP652), which exhibited potent METTL3 inhibition, favorable intravenous pharmacokinetic properties, and significant antitumor activity in preclinical models of both hematologic and solid malignancies. These findings also support the therapeutic potential of targeting METTL3 in cancer treatment.

Additionally, pyridine-based METTL3 inhibitors have also been independently pursued by Aqemia and HitGen (Table 2). Compound 28 from Aqemia demonstrated potent METTL3 inhibition (IC₅₀ < 10 nM) in MTase-Glo assays and showed antiproliferative effects in MOLM-13 and Kasumi-1 cell lines, with IC₅₀ values of 0.56 μM and 0.63 μM, respectively [56]. Similarly, Compound 29 from HitGen inhibited METTL3 enzymatic activity with IC₅₀ values below 100 nM and suppressed the growth of MOLM-13 cells at submicromolar concentrations [57]. Interestingly, both compounds possess electron-donating groups at the meta-position of the pyridine ring, a structural feature also found in compound 27, suggesting a structural motif that may enhance biological activity.

Table 2.

Chemical structures of compounds 28–29.

Compound	Structures
28
29

Despite the recent emergence of numerous highly potent substrate-competitive inhibitors, structural diversity of these compounds remains relatively limited. Furthermore, their cellular activity is often constrained by high intracellular levels of the endogenous substrate SAM. To overcome these challenges, alternative targeting strategies such as allosteric modulators and targeted protein degradation are being explored and may provide new opportunities to overcome current limitations in METTL3-targeted drug discovery. The next section highlights recent advances in these emerging therapeutic approaches targeting METTL3.

3.2. Allosteric inhibitors

A promising approach to targeting METTL3 involves the development of allosteric inhibitors that reversibly bind to regions outside the catalytic center of the METTL3-METTL14 complex. Unlike competitive inhibitors, these allosteric modulators retain their efficacy even under high intracellular concentrations of SAM. In 2022, Kim et al. reported the first such allosteric inhibitors against the METTL3-METTL14 complex [58]. Through high-throughput screening, they identified 30 (CIDBA), a previously reported cytosolic phospholipase A2 (cPLA2) inhibitor. Subsequent SAR optimization led to the development of compound 31, which inhibited METTL3-METTL14 enzymatic activity with an IC₅₀ of 2.8 μM (Table 3). A series of biochemical assays, including substrate competition studies and tests with various METTL3-METTL14 constructs, confirmed that 31 acts through allosteric binding at a site distant from the catalytic center.

Table 3.

Chemical structures of compounds 30–32.

Compound	Structures	IC50 (μM)
30		17.3
31		2.8
32		3.65

Additionally, this team also identified eltrombopag (compound 32) as another allosteric METTL3 inhibitor (Table 3) [59]. Eltrombopag demonstrated an IC₅₀ of 3.65 μM in enzyme-based assays and exhibited antiproliferative activity in the MOLM-13 cells. Substrate competition studies with varying concentrations of SAM and RNA substrates demonstrated that eltrombopag functions through a noncompetitive inhibition mechanism. Computational analysis, including cavity mapping and molecular docking, suggested that eltrombopag binds to a putative allosteric site of METTL3. Furthermore, eltrombopag treatment reduced m⁶A methylation levels and suppressed the expression of leukemogenic genes in MOLM-13 cells. These findings underscore the value of allosteric inhibitors as a compelling alternative strategy for modulating METTL3 activity, particularly in cellular contexts with abundant SAM.

3.3. Proteolysis targeting chimeras

Targeted protein degradation, exemplified by PROTACs, represents an emerging strategy in drug discovery [60,61]. PROTACs are heterobifunctional molecules that composed of a warhead targeting the protein of interest (POI), an E3 ubiquitin ligase ligand, and a linker. Upon formation of a POI-PROTAC-E3 ligase ternary complex, the POI is ubiquitinated and subsequently degraded via the ubiquitin-proteasome system (UPS). Unlike traditional inhibitors, PROTACs offer catalytic degradation of target proteins and can disrupt both enzymatic and scaffolding functions [62].

Given that METTL3 can exert oncogenic functions independent of its methyltransferase activity, PROTACs likely offer a promising strategy to comprehensively modulate its biological functions. In our prior work, we designed a series of PROTACs using UZH2—a potent and selective METTL3 inhibitor – as the POI-binding ligand [63]. By conjugating UZH2 at its methylamino position with various linkers to the von Hippel-Lindau (VHL) E3 ubiquitin ligase ligand, we synthesized a series of bifunctional PROTACs. Among these, compound 33 (WD6305) showed the most potent activity, inducing METTL3 degradation in a time- and dose-dependent manner, with a DC₅₀ of 140 nM and a D_max of 91.9% in Mono-Mac-6 cells (Table 4). WD6305 also facilitated the degradation of METTL14, likely by indirectly recruiting METTL14 to the E3 ligase for subsequent proteasomal degradation and/or by destabilizing the METTL3-METTL14 complex. Mechanistic studies confirmed that this degradation process is mediated by VHL E3 ligase and UPS. Functionally, WD6305 significantly outperformed UZH2 in inhibiting the proliferation and inducing apoptosis of AML cell lines, including Mono-Mac-6 and MOLM-16. Furthermore, WD6305 effectively reduced m⁶A RNA methylation levels and modulated gene expression profiles associated with AML pathogenesis. These findings position WD6305 as both a valuable tool for probing METTL3-METTL14 biology and a promising candidate for antileukemic therapy.

Table 4.

Chemical structures of compounds 33–37.

Compound	Structures	Activity
33 (WD6305)		DC50 = 140 nM Dmax = 91.9% (Mono-Mac-6)
34		50% degradation @ 2 μM (MOLM-13)
35 (KH12)		DC50 = 220 nM (MOLM-13)
36 (ZW27941)		DC50 = 170 nM (MOLM-13)
37 (AF151)		DC50 = 430 nM (MOLM-13)

Subsequent research by the Caflisch group also utilized UZH2 as the METTL3 ligand to develop a series of cereblon (CRBN)-based PROTACs. Among these, compound 34 demonstrated achieved over 50% degradation of METTL3 in MOLM-13 cells at 2 μM and exhibited antiproliferative activity across various AML cell lines and the PC3 prostate cancer cell lines [64]. In another study, Sim et al. developed compound 35 (KH12), a VHL-based PROTAC METTL3 degrader, to explore its non-enzymatic functions [65]. Compound 35 effectively inhibited the growth of multiple gastric cancer cell lines, whereas the UZH2 alone had minimal effects. These results align with previous findings that METTL3 promotes tumorigenesis through mechanisms beyond its catalytic function, and they support the hypothesis that PROTACs can effectively target these non-catalytic roles, leading to enhanced antiproliferative activity. Qian et al. also reported a VHL-based METTL3 PROTAC degrader, 36 (ZW27941) [66], which caused METTL3 degradation within 24 h with a DC₅₀ of 0.17 μM in MOLM-13 cells. Compared to small-molecule inhibitors, 36 demonstrated superior efficacy in inducing apoptosis and G₀/G₁ cell cycle arrest, and inhibiting colony formation. Furthermore, it exhibited synergistic or additive effects when combined with standard AML therapeutics such as cytarabine and venetoclax, suggesting its potential utility in combination therapy regimens.

In a very recent study, Barthels et al. reported the development of a series of novel METTL3-targeting degraders [67], utilizing compounds 15 (STM2457), 19 (STM3006), and 25 as POI ligands. Among these, only compound 37 (AF151), which incorporates a derivative of compound 25 as the POI-binding ligand, exhibited effective degradation activity, with a DC₅₀ of 430 nM in MOLM-13 cells. Moreover, compound 37 demonstrated enhanced biological activity compared to its parent inhibitors, including a lower IC₅₀ for cell viability (IC₅₀ = 0.45 µM) and a more pronounced reduction in m⁶A modification levels in cancer cell models. This work broadens the chemical space of METTL3-targeted PROTACs. Notably, all METTL3-targeting degraders discussed above (including compounds 33–37) consistently induce co-degradation of METTL14, indicating that this may be a characteristic feature of this class of compounds.

Collectively, these studies highlight the therapeutic promise of PROTAC-mediated degradation of METTL3 in cancers. The continued development of PROTACs provides new avenues for overcoming the limitations of traditional small-molecule inhibitors, especially for targeting non-catalytic protein functions.

4. Conclusion

Accumulating evidence supports METTL3 as a promising therapeutic target in cancer, owing to its pivotal role in regulating m⁶A RNA methylation and its impact on oncogenic transcriptional programs and tumor immune evasion. Pharmacological inhibition of METTL3 has demonstrated robust antitumor effects in a variety of preclinical or clinical cancer models, highlighting its potential for translational applications. Recent rapid progress has been made in the development of small-molecule modulators targeting METTL3, including substrate-competitive inhibitors, allosteric inhibitors, and proteolysis-targeting chimeras.

Among these, SAM-competitive inhibitors remain the most extensively explored, with diverse scaffolds such as aminopyrimidines, pyrimidinones and indazoles. Representative compounds, including 10 (UZH2) and 16 (STC-15), demonstrate potent biochemical and cellular activity, with STC-15 advancing to in vivo evaluation and early-phase clinical trials. However, their cellular efficacy is frequently attenuated due to the high intracellular concentration of SAM, which competes with the inhibitor for binding. Allosteric inhibitors, which engages sites outside the SAM-binding pocket, offer a promising strategy to overcome this limitation. While high-throughput screening has identified several candidate compounds, most exhibit only moderate biochemical potency (in the micromolar range), highlighting the need for further optimization.

In parallel, increasing recognition of METTL3’s non-catalytic roles in cancer progression has prompted interest in targeted protein degradation approaches. Current PROTAC strategies predominantly utilize UZH2 as the METTL3-binding ligand, with 33 (WD6305) emerging as one of the most effective degraders identified to date. Despite these advances, limited drug-like properties of current METTL3 PROTACs present challenges for evaluating in vivo efficacy. Therefore, continued efforts to develop novel, potent, and druggable METTL3 modulators will be essential to fully harness its therapeutic potential in cancer treatment.

5. Future perspective

Despite significant advances in development of modulators targeting METTL3, several challenges and opportunities remain in the field. Future efforts should focus on expanding the structural diversity of substrate-competitive inhibitors. Developing novel chemotypes with enhanced affinity, membrane permeability, and metabolic stability will be essential to address the limitations of this class of inhibitors. Simultaneously, allosteric inhibitors represent an attractive alternative for targeting METTL3. Identifying and optimizing ligands that bind allosteric sites – especially through emerging methodologies such as artificial intelligence-driven drug design (AIDD) and computer-aided drug discovery (CADD) – could yield compounds with improved selectivity and reduced susceptibility to SAM interference.

Another key direction involves the development of novel PROTACs targeting METTL3. Future PROTAC development should emphasize diversification of METTL3 ligands, E3 ligase binders, and linker architectures to optimize pharmacokinetics and drug-like properties. In vivo validation of such degraders will be crucial to advancing them toward clinical evaluation.

Additionally, the potential of METTL3-targeted modulators in tumor immunotherapy remains an area of significant interest and warrants further investigation. In addition to its direct tumor-intrinsic effects, METTL3 has been implicated in modulating immune cell function and facilitating tumor immune evasion. Therefore, combining small-molecule modulators of METTL3 with immune checkpoint inhibitors or other immunotherapies may elicit synergistic antitumor effects. Continued preclinical and clinical studies investigating these combination strategies could unlock new therapeutic opportunities, especially in immunotherapy-resistant malignancies.

In conclusion, advancing METTL3-targeted therapies will require sustained innovation across molecular design, mechanistic elucidation, and translational research. Realizing the full therapeutic potential of small molecules targeting the METTL3 will depend on integrating multidisciplinary approaches to overcome current limitations and enable broad application in cancer treatment.

Funding Statement

This manuscript was funded by the National Key Research and Development Program of China [2023YFA1800804].

Article highlights

METTL3’s Multifaceted Role in Tumor Biology

• METTL3 is overexpressed in various cancers, where it frequently functions as a tumor promoter through m⁶A-dependent mechanisms.

• Inhibition of METTL3 enhances antitumor immunity by remodeling the tumor microenvironment and activating interferon-mediated responses.

• Cytoplasmic METTL3 contributes to oncogenic translation independently of its methyltransferase activity.

Structure of METTL3-METTL14 Complex

• METTL3 serves as the sole catalytic subunit, while METTL14 plays a structural role in stabilizing the complex.

• S-adenosylmethionine (SAM) binds within the catalytic pocket of METTL3, inducing conformational changes in the gate loops.

• Most reported METTL3 inhibitors mimic the adenine moiety of SAM to competitively occupy the binding site.

Recent Progress in Small-molecule Modulators of METTL3

• METTL3 modulators fall into three primary categories: substrate-competitive inhibitors, allosteric inhibitors, and proteolysis-targeting chimeras (PROTACs).

• SAM-competitive inhibitors remain the most extensively explored, with representative scaffolds including aminopyrimidines, pyrimidinones, and indazoles.

• Current METTL3 PROTACs (e.g. WD6305) exhibit superior antiproliferative activity compared to their parent inhibitors.

Challenges and Future Perspective

• There remains a pressing need for novel chemotypes with enhanced binding affinity, membrane permeability, and metabolic stability.

• Artificial intelligence–driven drug design (AIDD) offers a promising strategy for identifying non-SAM-competitive compounds with improved selectivity.

• Broadening the diversity of METTL3 ligands and E3 ligase recruiters may enhance degradation efficiency and optimize drug-like properties.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing assistance

No writing assistance was utilized in the production of this manuscript.

Reviewer disclosure

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions statement

Feifei Wu: Investigation, Writing – Original Draft Preparation; Lei Yu: Conceptualization, Writing – Review & Editing; Shilin Xu: Conceptualization, Funding Acquisition, Supervision, Writing – Review & Editing.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the author(s) used ChatGPT to correct the grammatical and typographical errors in the manuscript. All authors have approved the final version of the manuscript.

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17568919.2025.2546781