snoRNA in cancer

snoRNAs, small nucleolar ncRNAs (60–300 nucleotides), are abundant in eukaryotic nucleoli (Ref. Reference Kiss8). They play dual roles in cancer progression (Ref. Reference Williams and Farzaneh9). Tumor-suppressive snoRNAs like SNORD50A/B has been shown to inhibit BS (BC) by binding K-RAS (Ref. Reference Siprashvili, Webster, Johnston, Shenoy, Ungewickell, Bhaduri, Flockhart, Zarnegar, Che, Meschi, Puglisi and Khavari10). Similarly, SNORD113–1 inhibits tumorigenesis by regulating MAPK/ERK and TGF-β signaling pathways (Ref. Reference Xu, Yang, Ding, Zhao, Ren, Zhao, Wang and Qi11). Conversely, oncogenic snoRNAs such as SNORD17 promote hepatocellular carcinoma (HCC) through a positive feedback loop involving nucleolar phosphoprotein 1 (NPM1) and human Myb-binding protein 1A (MYBBP1A) upon p53 inactivation (Ref. Reference Liang, Li, Liao, Huang, Wen, Wang, Chen, Cai, Xu, Ding, Liang, Datta, Chu, Chen and Zhang12), and SNORA23 promotes tumor cell proliferation and invasion by upregulating the expression of spectrin repeat containing nuclear envelope 2 (SYNE2) (Ref. Reference Cui, Nakano, Obchoei, Setoguchi, Matsumoto, Yamamoto, Obika, Shimada and Hiraoka13). Consequently, snoRNAs are implicated in tumorigenesis and may emerge as pivotal biomarkers for diagnosing and predicting the outcome of various cancers.

rRNA in cancer

rRNAs affect cancer progression through gene expression regulation (Ref. Reference Barbieri and Kouzarides6), ribosome assembly modulation (Ref. Reference Bai, Zhang, Li, Hang, Liu, Tian, Huang, Qu, Cao, Ji and Zheng14), and oncogenic protein translation (driving proliferation/transformation). Their roles as therapeutic targets and prognostic markers (Ref. Reference Liu, Liu, Liu, Wu, Lin, Liu, Zhang, Wang, Prager, Gimple, Yu, Zhao, Wu, Zhang, Wu, Chen, Luo, Rich, Xie, Jiang and Chen15), highlight novel strategies for cancer diagnosis, treatment, and outcome prediction, offering innovative insights into cancer management.

miRNA in cancer

miRNAs, short RNAs (19–25 nucleotides), regulate gene expression and exhibit dual roles in cancer as oncogenes or tumor suppressors (Ref. Reference Lu and Rothenberg16). A foundational study revealed reduced miR-15/miR-16 expression in chronic lymphocytic leukemia (CLL) (Ref. Reference Calin, Dumitru, Shimizu, Bichi, Zupo, Noch, Aldler, Rattan, Keating, Rai, Rassenti, Kipps, Negrini, Bullrich and Croce17), while miR-31 promotes endometrial cancer (EC) by inhibition of the Hippo pathway (Ref. Reference Mitamura, Watari, Wang, Kanno, Kitagawa, Hassan, Kimura, Tanino, Nishihara, Tanaka and Sakuragi18). The miR-34 family, regulated by tumor suppressor p53, controls cell growth, apoptosis, cell cycle (Ref. Reference Hermeking19). Notably, miR-34a loss has been shown to promote colorectal cancer (CRC) and predict poor survival in CRC patients via activating the IL6-STAT3 signaling pathway (Ref. Reference Liu, Rokavec, Huang and Hermeking20). Importantly, miR-34a may serve as the most promising miRNA drugs for cancer treatment (Ref. Reference Kim and Croce21), underscoring diagnostic and therapeutic value of miRNAs in cancer management. Hence, miRNAs broadly influence cancer progression by modulating cell growth, invasion/metastasis, angiogenesis, and cellular transformation (Refs Reference Slack and Chinnaiyan5, Reference Lee and Dutta22).

lncRNA in cancer

lncRNAs, transcripts exceeding 200 nucleotides (Ref. Reference Mattick, Amaral, Carninci, Carpenter, Chang, Chen, Chen, Dean, Dinger, Fitzgerald, Gingeras, Guttman, Hirose, Huarte, Johnson, Kanduri, Kapranov, Lawrence, Lee, Mendell, Mercer, Moore, Nakagawa, Rinn, Spector, Ulitsky, Wan, Wilusz and Wu23), regulate cancer development by modulating proliferation, differentiation, and metastasis (Refs Reference Li, Wang, Zhu and Wang24, Reference Li, Shu, Geng, Ying, Guo, Luo, Xiang, Wu, Ma, Chan, Zhu, Ambinder and Tao25 Reference Zangouei, Zangoue, Taghehchian, Zangooie, Rahimi, Saburi, Alavi and Moghbeli26, Reference Ji, Diederichs, Wang, Böing, Metzger, Schneider, Tidow, Brandt, Buerger, Bulk, Thomas, Berdel, Serve and Müller-Tidow27). Oncogenic lncRNAs include prostate cancer antigen 3 (PCA3), elevated in prostate cancer (PCa) patients’ urine as an early diagnostic biomarker (Ref. Reference Chevli, Duff, Walter, Yu, Capuder, Elshafei, Malczewski, Kattan and Jones28), and lncRNA programmed cell death ligand 1 (PD-L1), which accelerates lung adenocarcinoma (LUAD) by enhancing c-Myc transcriptional activity (Ref. Reference Qu, Jiao, Lu, Yao, Wang, Rong, Xu, Fan, Sun, Yang, Wang, Yao, Xu, Yan, Wang, Liang and Zen29). Furthermore, lncRNA KIMAT1 may represent a therapeutic target for KRAS-driven lung cancer (LC) (Ref. Reference Shi, Magee, Fassan, Sahoo, Leong, Lee, Sellers, Brullé-Soumaré, Cairo, Monteverde, Volinia, Smith, Di Leva, Galuppini, Paliouras, Zeng, O’Keefe and Garofalo30). Conversely, tumor-suppressive lncRNAs like BC069792 inhibit BC by sponging miR-658 and miR-4739 as a competitive endogenous RNA (ceRNA), upregulating KCNQ4, and suppressing AKT phosphorylation to block metastasis (Ref. Reference Zhang, Dong, Guo, Li, Fan, Liu, Yuan, Ma, Wang, Zheng, Li and Gao31). These findings underscore the pivotal function of lncRNAs in gene regulatory networks, indicating their promise as trustworthy diagnostic indicators or targets for cancer treatment (Ref. Reference Tan, Lin, Li, Li, Xu and Ju32).

siRNA in cancer

siRNAs regulate eukaryotic genome expression and function by modulating endogenous genes and protecting the genome against invading nucleic acids (Ref. Reference Carthew and Sontheimer33). Their primary function is in RNA interference (RNAi), a highly specific regulation that governs gene expression in a base-pairing manner (Ref. Reference Elbashir, Harborth, Lendeckel, Yalcin, Weber and Tuschl34). They modulate tumor-related signaling pathways (Refs Reference Elbashir, Harborth, Lendeckel, Yalcin, Weber and Tuschl34, Reference Jiang, Qi, Yang, Miao, Ren, Liu, Huang, Huang, Chen, Shi and Cai35), and offer targeted cancer therapy potential by silencing oncogenes with low doses, minimal side effects, highlighting their promise in precision oncology (Ref. Reference El Moukhtari, Garbayo, Amundarain, Pascual-Gil, Carrasco-León, Prosper, Agirre and Blanco-Prieto36).

piRNA in cancer

piRNAs (24–31 nucleotides) exert regulatory effects by interacting with Piwi proteins (Ref. Reference Liu, Dou, Song, Dong, Liu, Liu, Tao, Li, Yin and Xu37). Oncogenic piR-823 enhances DNA methylation in multiple myeloma (MM) (Ref. Reference Ai, Mu, Sun, Fan, Yan, Qin, Cui, Wang, Guo, Mei, Wang and Hu38), while piRNA-14633 drives cervical cancer (CC) via the methyltransferase-like protein 14 (METTL14)/CYP1B1 signaling axis (Ref. Reference Xie, Li, Luo, Wang, Zhou, Zhao, Gao, Yang, Fu, Kong and Sun39). Tumor-suppressive piR-39980 targets ribonucleotide reductase subunit M2 (RRM2) (Ref. Reference Das, Roy, Jain and Mallick40), and piR-36712 downregulation results in elevated selenoprotein W pseudogene 1 (SEPW1), which may inhibit p53, upregulating Slug to promote cell proliferation, invasion, and migration (Ref. Reference Tan, Mai, Zhang, Jiang, Zhang, Bai, Ye, Li, Pan, Su, Zheng, Liu, Zuo, Zhao, Li, Huang, Yang, Tan, Zheng and Lin41). Their dual roles, mediated through upstream events (e.g., methylation, gene silencing), underscore their impact on tumorigenesis. Their involvement in critical regulatory networks positions piRNAs as promising biomarkers for early cancer detection and therapeutic targets, offering the potential for precision interventions in oncology.

circRNA in cancer

circRNAs, covalently closed loops formed via precursor mRNA (pre-mRNA) back-splicing (Ref. Reference Chen42), influence cancer progression by modulating metastasis and invasiveness (Refs Reference Zhao, Lee, Kim, Yang, Chamseddin, Ni, Gusho, Xie, Chiang, Buszczak, Zhan, Laimins and Wang43, Reference Xu, Wang, Tang, Feng, Zhang, Liu, Wu, Wang, Feng, Tang, Lisch and Lu44). For instance, hsa_circ_0003258 promotes PCa metastasis by upregulating Rho GTPase-activating protein 5 (ARHGAP5) expression, stabilizing histone deacetylase 4 (HDAC4) mRNA via insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) binding, activating of the ERK signaling pathway, and triggering epithelial-mesenchymal transition (EMT) (Ref. Reference Yu, Lv, Wang, Song, Xie, Wang, Li, Guo, Fu, Li, Wu, Chan, Feng, Chen and Zhao45). Similarly, circSATB2 drives non-small cell lung cancer (NSCLC) progression via the miR-326/FSCN1 axis and facilitates exosome-mediated intercellular communication, highlighting its diagnostic potential (Ref. Reference Zhang, Nan, Chen, Li, Jia, Qiu, Dai, Zhou, Zhu, Zhang and Jiang46). Their aberrant expression across various cancers positions them as promising biomarkers for early detection and therapeutic targets (Refs Reference Chen, Huang, Xu, Livingstone, Soares, Jeon, Zeng, Hua, Petricca, Guo, Wang, Yousif, Zhang, Donmez, Ahmed, Volik, Lapuk, Chua, Heisler, Foucal, Fox, Fraser, Bhandari, Shiah, Guan, Li, Orain, Picard, Hovington, Bergeron, Lacombe, Fradet, Têtu, Liu, Feng, Wu, Shao, Komor, Sahinalp, Collins, Hoogstrate, de Jong, Fijneman, Fei, Jenster, van der Kwast, Bristow, Boutros and He47, Reference Pisignano, Michael, Visal, Pirlog, Ladomery and Calin48).

N6-methyladenosine (m6A)

N6-methyladenosine (m6A), the most prevalent and distinctive form of RNA methylation modification (Ref. Reference Barbieri and Kouzarides6), involves reversible methyl group addition by ‘Writers’ (methyltransferases), removal by ‘Erasers’ (demethylases), and recognition by ‘Readers’ (m6A-binding proteins) (Figure 1a), critically regulating RNA metabolism and function (Refs Reference Nombela, Miguel-López and Blanco49, Reference Akichika, Hirano, Shichino, Suzuki, Nishimasu, Ishitani, Sugita, Hirose, Iwasaki, Nureki and Suzuki50). This dynamic modification influences gene expression in physiological and pathological conditions, including cancer (Ref. Reference He, Li, Wu, Peng, Shu and Yin51). For instance, methyltransferase-like protein 3 (METTL3)- mediated m6A facilitates miRNA maturation by guiding DiGeorge syndrome critical region 8 (DGCR8) to primary miRNA (pri-miRNA) (Ref. Reference Alarcón, Lee, Goodarzi, Halberg and Tavazoie52). Cigarette smoke condensate (CSC) may be a promoter of pri-miR-25 maturation through METTL3-mediated m6A modification, activating the AKT-p70S6K signaling pathway and potentially promoting cancer development (Ref. Reference Zhang, Bai, Li, Ye, Wu, Wang, Li, Tan, Mai, Li, Pan, Zheng, Su, Ye, Fu, Zheng, Zuo, Liu, Zhao, Che, Xie, Jia, Zeng, Tan, Chen, Xu, Zheng and Lin53). m6A also drives circRNA biogenesis and alters lncRNA stability (Ref. Reference Liu, Gao, Cheng, Lv, Sun, Wang and Tang7). However, m6A dysregulation may disrupt ncRNA stability, localization, and function, thereby affecting the regulation of gene expression and promoting the proliferation, invasion and metastasis of tumor cells (Ref. Reference Yang, Liu, Li, Zhang, Hiju, Sun, Mao, Zheng and Feng54). Collectively, m6A epitranscriptomic regulation bridges RNA modification with disease etiology, highlighting its potential as a biomarker and intervention target.

Figure 1. Chemical modifications of RNAs and their main functions. (A) m6A modification regulates the stability, translation, alternative splicing and nuclear export of RNAs mediated by writers, including METTL3-methyltransferase-like 14 (METTL14), methyltransferase-like 16 (METTL16), Zinc Finger CCHC-Type Containing 4 (ZCCHC4), erasers FTO and α-ketoglutarate-dependent dioxygenasehuman AlkB homolog 5 (ALKBH5), and reader proteins YTH structural domain family proteins 1–3 (YTHDF1–3) and YTH structural domain-containing proteins 1–2 (YTHDC1–2). (B) m6Am modification regulates the stability of RNAs mediated by writers, including METTL3-METTL14 and erasers FTO. (C) m1A modification regulates the decay of RNAs mediated by writers, including tRNA methyltransferase 6 (TRMT6) and tRNA methyltransferase 61A (TRMT61A), erasers ALKBH3 and reader proteins YTHDF2/3. (D) m5C modification regulates the nuclear export of RNAs mediated by writers NOL1/NOP2/SUN domain family member (NSUN) 2/6 and reader proteins Aly/REF export factor (ALYREF). 5hmC is formed from 5mC by oxidation of 10–11 translocation (TET) proteins, regulating the stability, translation, alternative splicing and nuclear export of RNAs. (E) ac4C modification regulates the stability and translation of RNAs mediated by writers N-acetyltransferase 10 (NAT10). (F) m7G modification regulates the stability of RNAs mediated by writers methyltransferase-like protein-1 (METTL1). (G) Nm regulates the stability of RNAs mediated by writers, including FtsJ homolog 3 (FTSJ3), rRNA 2’-O-methyltransferase fibrillarin (FBL) and tRNA methyltransferase (Trm) 7/13/56/J. (H) Ψ modification regulates the processing and translation of RNAs mediated by writers, including DKC1, PUS, probable tRNA pseudouridine synthase 1 (TRUB1) and RNA pseudouridylate synthase domain containing 4 (RPUSD4). (I) Glycosylated RNAs act as ligands in immunoregulation under the regulation of glycosltransferases (GTFs).

N6,2’-O-dimethyladenosine (m6Am)

m6Am, the methylation of adenosine’s nitrogen at position 6 (Ref. Reference Wei, Gershowitz and Moss55), modulates the stability of ncRNAs (Figure 1b) (Ref. Reference Nombela, Miguel-López and Blanco49), thereby influencing disease development (Ref. Reference Jia, Chai, Wang, Sun, Xu, Yang, Ge, Jia, Yang and Fan56). Though research is limited, further studies may reveal novel therapeutic avenues, particularly in metabolic and immunotherapeutic applications.

N1-methyladenosine (m1A)

m1A, the N1-methylation of adenine, influences immune responses by promoting T-cell expansion through tRNA modification (Figure 1c) (Ref. Reference Liu, Zhou, Li, Zhang, Shi, Wang, Li, Miao, Chen, He, Dong, Lee, Zheng, Liu, Su, Ye, Flavell, Yi, Wu and Li57). Detected via methylated RNA immunoprecipitation sequencing (MeRIP-seq) (Ref. Reference Dominissini, Nachtergaele, Moshitch-Moshkovitz, Peer, Kol, Ben-Haim, Dai, Di Segni, Salmon-Divon, Clark, Zheng, Pan, Solomon, Eyal, Hershkovitz, Han, Doré, Amariglio, Rechavi and He58), advancements in high-throughput sequencing now enable precise localization and quantification of m1A modifications, revealing their roles in ncRNAs (Refs Reference Dominissini, Nachtergaele, Moshitch-Moshkovitz, Peer, Kol, Ben-Haim, Dai, Di Segni, Salmon-Divon, Clark, Zheng, Pan, Solomon, Eyal, Hershkovitz, Han, Doré, Amariglio, Rechavi and He58-Reference Liu, Clark, Luo, Wang, Fu, Wei, Wang, Hao, Dai, Zheng, Ma, Han, Evans, Klungland, Pan and He60). Further research could deepen understanding of its biological functions, offering insights into ncRNA mechanisms and therapeutic potential.

5-methylcytosine (m5C)

m5C, the methylation at the fifth carbon of cytosine, represents a crucial RNA epigenetic modification regulating gene expression by influencing RNA stability and nuclear export (Figure 1d) (Ref. Reference Bohnsack, Höbartner and Bohnsack61). In tRNAs, m5C enhances translation efficiency through optimized codon-anticodon pairing on tRNAs and mRNAs (Ref. Reference Song, Zhang, Liu, Xu, Cai, Yang, Straube, Yu and Ma62). Moreover, m5C is essential for cell functions like stress responses and metabolic processes (Ref. Reference Li, Tao, Zhao, Li, Zheng, Li and Chen63). m5C alterations in ncRNAs can affect tumor progression (Refs Reference Song, Zhang, Liu, Xu, Cai, Yang, Straube, Yu and Ma62, Reference Li, Tao, Zhao, Li, Zheng, Li and Chen63, Reference Tang, Li, Wang, Wu, Hou, Ouyang, Fan, Li, Wang, Guo, Zhou, Liao, Wang, Xiang, Jiang, Li, Zeng and Xiong64), suggesting their potential clinical value.

N4-acetylcytidine (ac4C)

ac4C, the acetylation of cytosine’s fourth nitrogen, ensures translation accuracy and was initially identified in tRNAs (Refs Reference Stern and Schulman65, Reference Liu, Zheng, Li, Fei, Wang, Zhang, Wang, Jiang and Wang66). It also exists in rRNAs and mRNAs, affecting RNA stability and function (Figure 1e) (Refs Reference Chen, Hao, Liu, Zhong, You, Ao, Chong, Luo, Yin, Ye, He, Lu, Chen, Li, Zhang and Guo67, Reference Liu, Wubulikasimu, Cai, Meng, Cui, Zhou and Li68, Reference Yan, Zhou, Wang, Ding, Ma, Li, Jia, Gao and Lu69). Linked to diseases like cancer, ac4C abnormalities in ncRNAs highlight their potential as diagnostic biomarkers and therapeutic targets.

N7-methylguanosine (m7G)

m7G, the methylation at guanosine’s 7th nitrogen, occurs in RNAs like mRNA and tRNA, regulating post-transcriptional processing, stability, translation, degradation, and interactions with RNA-binding proteins (RBPs) (Figure 1f) (Refs Reference Pandolfini, Barbieri, Bannister, Hendrick, Andrews, Webster, Murat, Mach, Brandi, Robson, Migliori, Alendar, d’Onofrio, Balasubramanian and Kouzarides70, Reference Zhang, Liu, Ma, Dai, Sun, Luo, Zhang, Zhang, Hu, Dong and He71, Reference Xia, Wang and Zheng72, Reference Zhang, Zhu, Shen, Shen and Chen73). Additionally, tools like m7GDisAI have been established to identify potential disease-related m7G loci (Ref. Reference Ma, Zhang, Chen, Song, Zang and Liu74), aiding research into its roles in RNA biology and potential therapeutic applications.

2’-O-methylation (2’-O-Me, Nm)

Nm modification involves methylating RNA ribose’s 2′-hydroxyl group, stabilizing piRNAs, maintaining tRNA function, protecting mRNAs from decapping exoribonuclease (DXO) degradation, and ensuring rRNA biogenesis(Figure 1g) (Ref. Reference Hassan, Ariyur, Daulatabad, Mir and Janga75). Like m6A, Nm modification is increasingly studied for its diverse functions and regulatory mechanisms in ncRNAs (Ref. Reference Sun, Li, Liu and Yi76), underscoring its biological importance and potential as a therapeutic target.

Pseudouridine (Ψ)

Pseudouridine (Ψ), termed the “fifth nucleoside” of RNA, is formed by pseudouridine synthases (PUSs) through a β-glycosidic bond linking uracil’s C-5 to ribose’s C-1, creating a structural isomer of uridine (Refs Reference Li, Ma and Yi77, Reference Rodell, Robalin and Martinez78). Abundant in ncRNAs like tRNAs and snRNAs, Ψ modifications are involved in cellular activities and contribute to pathological conditions (Refs Reference Song, Zhuang, Zhu, Meng, Lu, Xie, Peng, Li and Yi79, Reference Zhang, Jiang, Ma, Liu, Zhuang, Lu, Li, Peng and Yi80). For example, PUS10 regulates nuclear miRNA processing and cytoplasmic tRNA pseudouridylation (Ref. Reference Song, Zhuang, Zhu, Meng, Lu, Xie, Peng, Li and Yi79), while PUS7 overexpression correlates with poor prognosis in patients with glioblastoma (GBM) by enhancing tyrosine kinase 2 (TYK2) translation efficiency of via tRNA pseudouridylation (Figure 1h), thereby promoting glioblastoma stem cell (GSC) growth (Ref. Reference Cui, Yin, Zhang, Ye, Chen, Chao, Meng, Wei, Roeth, Li, Qin, Sun, Zhang, Klein, Huynhle, Wang, Zhang, Badie, Kalkum, He, Yi and Shi81). The dyskerin pseudouridine synthase 1 (DKC1) gene encodes a dyskerin protein with PUS activity, which binds to and catalyzes the uridine isomerization of target RNAs to Ψ. Cancer progression and poor prognosis are linked to the overexpression of DKC1, which has been detected in a range of cancer types (Ref. Reference Barbieri and Kouzarides6, Reference Kan, Wang, Sheng, Chen, Yao, Mao and Chen82). However, the specific function of Ψ-modified ncRNAs in cancer biology is yet to be fully understood.

Glycosylation

Glycosylation, the enzymatic addition of sugar residues to proteins or lipids in the endoplasmic reticulum and Golgi apparatus, is closely linked to cancer progression by influencing tumor growth, invasiveness, and immune evasion (Figure 1i) (Ref. Reference Bangarh, Khatana, Kaur, Sharma, Kaushal, Siwal, Tuli, Dhama, Thakur, Saini and Saini83). Traditionally associated with proteins and lipids, this paradigm was challenged when Flynn et al. discovered glycosylated small non-coding RNAs (sncRNAs) produced via the classic protein N-glycosylation pathway. These glycosylated sncRNAs, found on cell surfaces, interact with Siglec receptors to regulate immune responses (Ref. Reference Flynn, Pedram, Malaker, Batista, Smith, Johnson, George, Majzoub, Villalta, Carette and Bertozzi84). Glycosylation also modifies Ψ in tRNA anticodons, critical for post-embryonic growth by maintaining codon translation and protein stability (Ref. Reference Zhao, Ma, Ishiguro, Saito, Akichika, Matsuzawa, Mito, Irie, Ishibashi, Wakabayashi, Sakaguchi, Yokoyama, Mishima, Shirouzu, Iwasaki, Suzuki and Suzuki85). In situ imaging reveals dynamic glycosylated RNA levels during disease and physiological processes: they increase during pro-inflammatory monocyte/macrophage-vascular endothelial cell interactions and decrease during immune differentiation and BC metastasis (Ref. Reference Ma, Guo, Mou, Shao, Lyu, Garcia, Kong, Lewis, Ward, Yang, Pan, Yi and Lu86). Cell-surface glycosylated RNAs are recognized by endothelial P-selectin, implicating them in neutrophil-mediated inflammation and tumor development (Ref. Reference Zhang, Tang, Torres, Wang, Ajaj, Zhu, Luan, Zhou, Wang, Zhang, Kurbatov, Khan, Kumar, Hidalgo, Wu and Lu87). These findings underscore glycosylation’s expanded role beyond classical substrates, highlighting its regulatory functions in RNA biology, immune modulation, and disease mechanisms. Glycosylated RNAs may serve as novel biomarkers or therapeutic targets in cancer and inflammatory disorders, though further research is needed to unravel their precise molecular roles and clinical potential.

Chemically modified ncRNAs in cancer occurrence and development

RNA modifications play critical roles in the development of cancer through diverse mechanisms. For instance, Li et al. highlighted the impact of m6A modification in super-enhancer RNA (seRNA) on histone modification and oncogene expression in pancreatic ductal adenocarcinoma (PDAC) (Ref. Reference Li, Zhao, Huang, Zhang, Bai, Zhuang, Wen, Wu, Zhou, Li, Zeng, Zhang, Deng, Su, Zuo, Chen, Lin and Zheng88). Additionally, the upregulation of lncRNA LCAT3, facilitated by m6A modification, promotes the growth, migration as well as invasiveness of LC cell via LCAT3-FUBP1-cMYC axis, leading to a poor prognosis (Ref. Reference Qian, Yang, Qiu, Li, Jiang, Li, Dong, Ying, Lu, Chen, Liu and Lu89). Conversely, circGPR137B is identified as a cytoplasmic sponge for miR-4739, which in turn upregulates fat mass and obesity-associated protein (FTO)’ expression. The demethylation of circGPR137B by FTO, which targets m6A, has been found to inhibit cell growth, thereby suppressing HCC metastasis (Ref. Reference Liu, Gu, Ma, Wang, Li, Wang, Yin and Li90). METTL3 hypomethylation, induced by CSC via the transcription factor NFIC, enhances m6A methylation of pri-miR-25, accelerating its maturation via NF-κB activating protein (NKAP)-Drosha-DGCR8 complexes to activate oncogenic AKT signaling in the initiation and progression of PDAC (Reference Zhang, Bai, Li, Ye, Wu, Wang, Li, Tan, Mai, Li, Pan, Zheng, Su, Ye, Fu, Zheng, Zuo, Liu, Zhao, Che, Xie, Jia, Zeng, Tan, Chen, Xu, Zheng and Lin53). Yang et al. also found that METTL14, by forming a complex with Wilms’ tumor 1-associated protein (WTAP), mediates m6A modification of lncRNA XIST, leading to its degradation by YTH structural domain family proteins 2 (YTHDF2) and subsequently inhibiting CRC proliferation and metastasis (Ref. Reference Yang, Zhang, He, Xue, Zhang, He, Zang, Feng, Sun and Zheng91). Moreover, m5C methyltransferases drive oncogenesis by catalyzing m5C modifications of target RNAs (Refs Reference Wang, Yuan, Wang, Yang, Yuan, Ye, Yang, Zhou, Li, Li and Sun92, Reference Li, Li, Luo, Zhou, Shi, Yang, Lu, Chen, Sun and He93). In gastric cancer (GC), m5C modification of lncRNA NR_033928 is associated with its upregulation in cells and tissues, affecting cell growth and apoptosis (Ref. Reference Fang, Huang, Lv, Chen, Lu, Jiang, Xu, Li, Wang, Li, Li, Wang and Xu94). The specific disruption of circE7 in CC cells has been demonstrated to lead to decreased E7 protein levels and inhibited cancer cell growth (Ref. Reference Zhao, Lee, Kim, Yang, Chamseddin, Ni, Gusho, Xie, Chiang, Buszczak, Zhan, Laimins and Wang43). The deletion of methyltransferase-like protein-1 (METTL1) and WD repeat domain 4 protein (WDR4) impairs m7G modification of tRNA, reducing the growth, colony formation, and invasiveness of LC cells (Ref. Reference Ma, Han, Huang, Yang, Zheng, Cai, Bi, Huang, Liu, Huang, Luo, Li and Lin95). Additionally, METTL1 was found to influence the stability and maturation of let-7e by m7G modification at the G11 site, thereby inhibiting LC progression (Ref. Reference Pandolfini, Barbieri, Bannister, Hendrick, Andrews, Webster, Murat, Mach, Brandi, Robson, Migliori, Alendar, d’Onofrio, Balasubramanian and Kouzarides70). SNORD88C was identified as an oncogenic snoRNA that mediates Nm modification of 28S rRNA, affecting the translation of stearoyl-CoA desaturase1 (SCD1) and inhibiting cellular autophagy, thereby promoting the metastasis of NSCLC (Ref. Reference Wang, Wang, Zhang, Xie, Song and Song96). Similarly, lncRNA ZFAS1 recruits NOP58 to mediate SNORD12C/78-dependent Nm modification, stabilizing rRNA and regulating downstream genes to control cancer cell proliferation and apoptosis (Ref. Reference Wu, Qin, Lu, Wang, Zhang, Sun, Hu, Li, Chen, Wang, Zhao, Piao, Zhang and Wei97). These findings underscore RNA modifications as central regulators of oncogenic pathways, offering potential therapeutic targets. The interplay between RNA modifications, non-coding RNAs, and protein complexes highlights their multifaceted roles in cancer biology, emphasizing the need for further research to translate these insights into clinical strategies.

Chemically modified ncRNAs in cancer diagnosis and prognosis

RNA modifications offer promising avenues for addressing limitations in current cancer biomarkers, which often lack specificity and sensitivity (Refs Reference Konno, Koseki, Asai, Yamagata, Shimamura, Motooka, Okuzaki, Kawamoto, Mizushima, Eguchi, Takiguchi, Satoh, Mimori, Ochiya, Doki, Ofusa, Mori and Ishii98, Reference Delaunay, Pascual, Feng, Klann, Behm, Hotz-Wagenblatt, Richter, Zaoui, Herpel, Münch, Dietmann, Hess, Benitah and Frye99). In CRC, SNORA56-driven pseudouridylation (Ψ) of 28S rRNA promotes cell proliferation and correlates with poorer 5-year survival, suggesting it as a prognostic biomarker (Ref. Reference Xu, Bian, Wang, Niu, Liu, Xiao, Zhu, Huang, Zhang, Chen, Wu, Sun, Zhu and Pan100). Similarly, circ1662 (Ref. Reference Chen, Yuan, Zhou, Shao, Guo, Wang, Yang, Guo, Zhao, Dang, Yang, Wang, Kang, Ji, Liu and Sun101) and m6A-modified lncRNA RP11 (Ref. Reference Wu, Yang, Chen, Tian, Jiang, Chen, Li, An, Lu, Luo, Du, Shan, Liu and Wang102) – upregulated by zinc-finger E-box binding protein 1 (ZEB1) – emerge as diagnostic and prognostic markers for CRC metastasis, while m6A-altered circNSUN2 is linked to liver metastasis (Ref. Reference He, Jiang and Luo103). In HCC, m6A-modified miRNAs show superior diagnostic accuracy over traditional biomarkers like AFP for early detection (Reference Zhang, Chen, Tao, Yi, Lin, Li, Shao, Lin and Chen104), and SNORA24-directed Ψ modifications influence translational fidelity (Ref. Reference McMahon, Contreras, Holm, Uechi, Forester, Pang, Jackson, Calvert, Chen, Quigley, Luk, Kelley, Gordan, Gill, Blanchard and Ruggero105), suggesting utility in predicting therapeutic responses. Furthermore, a comprehensive analysis of m6A-associated lncRNAs in HCC has shed light on their potential mechanisms in regulating the immune microenvironment, offering new insights into their role and prognostic value in the disease (Ref. Reference Dai, Li, Ye, Yu, Deng, Liu, Li, Yang and Wang106). In the context of ovarian cancer (OC), the expression of ALKBH5 has been associated with resistance to platinum-based chemotherapy (Ref. Reference Liu, Zhang, Wang and Wang107), suggesting that m6A modification and its regulators may serve as potential biomarkers for the diagnosis of cancer. Current studies emphasize RNA modifications’ dual roles as disease drivers and biomarkers, urging further research to standardize detection methods. Integrating these modifications into existing diagnostic frameworks could enhance early detection, and improve outcomes across diverse cancers.

Chemically modified ncRNAs in cancer therapy

Aberrant RNA modifications are emerging as promising therapeutic targets in cancer, with inhibitors and immunotherapies showing preclinical efficacy (Refs Reference Xue, Chu, Zheng, Jiang, Bao, Su, Lu and Li108, Reference De Mey, Esprit, Thielemans, Breckpot and Franceschini109, Reference Zhang, Xu and Su110). The METTL3 inhibitor STM2457 suppresses growth, invasiveness, and migration of intrahepatic cholangiocarcinoma (ICC) cells, induces apoptosis, and triggers cell cycle arrest, thereby significantly suppressing ICC progression and exhibiting superior anti-tumor effects (Ref. Reference Xu, Tien, Shi, Chen, Zhu, Huang, Huang, Zhao and Yin111). FTO has been identified as a significant contributor to cancer cell growth and evasion of immune responses (Refs Reference Liu, Liang, Xu, Dong, Dong, Qiu, Zhang, Li, Huang, Li, Wu, Yin, Zhang, Guo, Liu, Xi, Jiang, Han, Yang and Xu112, Reference Chen, Zhang, Sze, KM-F, Huang, Wang, Lee, Lu, Lyu, Lee, M-FJ, Ho DW-H and Ng113, Reference Li, Su, Deng, Chen and Chen114). The utilization of small molecule FTO inhibitors, such as CS1 and CS2, has shown promising anti-cancer properties by directly interacting with FTO’s catalytic site. These inhibitors effectively suppress the demethylating function of FTO, hinder its binding to target mRNAs, and show significant inhibitory effects on various cancers like breast and pancreatic cancers (Ref. Reference Su, Dong, Li, Gao, Han, Wunderlich, Deng, Li, Huang, Gao, Li, Zhao, Robinson, Tan, Qing, Qin, Prince, Xie, Qin, Li, Shen, Sun, Kulkarni, Weng, Huang, Chen, Zhang, Wu, Olsen, Müschen, Marcucci, Salgia, Li, Fathi, Li, Mulloy, Wei, Horne and Chen115). Moreover, FTO promotes tumor cell glycolytic metabolism through epitranscriptomic regulation, leading to T-cell suppression and induced tumor immune evasion. The combination of PD-L1 blockade and the FTO inhibitor Dac51 has demonstrated enhanced tumor growth inhibition and accelerated activation of CD8⁺ T cells, resulting in improved tumor cell eradication (Ref. Reference Liu, Liang, Xu, Dong, Dong, Qiu, Zhang, Li, Huang, Li, Wu, Yin, Zhang, Guo, Liu, Xi, Jiang, Han, Yang and Xu112). Additionally, in T-cell acute lymphoblastic leukemia (T-ALL), the IGF2BP2 inhibitor JX5 disrupts NOTCH1 receptor’s mRNA stability, overcoming chemoresistance (Ref. Reference Feng, Chen, Wang, Li, Li, Li, Zhang, Li, Zhang, Ye, Zhao, Li and Ji116). Furthermore, the overexpression of YTH structural domain family proteins 1 (YTHDF1) in CRC has been correlated with CRC metastasis, implying that YTHDF1-m6A-ARHGEF2 may be a promising target for therapeutic intervention (Ref. Reference Wang, Gao, Zeng, Zhu, Mo, Wong, Bao, Su, Zhai, Wang, Soares, Xu, Chen, Hezaveh, Ci, He, McGaha, O’Brien, Rottapel, Kang, Wu, Zheng, Cai and Yu117). Studies have also highlighted the role of ac4C modification by N-acetyltransferase 10 (NAT10) in promoting the development of CC, with immunotherapy targeting NAT10 showing a synergistic effect with PD-L1 blockade (Ref. Reference Chen, Hao, Liu, Zhong, You, Ao, Chong, Luo, Yin, Ye, He, Lu, Chen, Li, Zhang and Guo67). C57, a PUS7 inhibitor, can effectively inhibit cell growth of GSC, and prolong the survival of mice with glioblastoma burden (Ref. Reference Cui, Yin, Zhang, Ye, Chen, Chao, Meng, Wei, Roeth, Li, Qin, Sun, Zhang, Klein, Huynhle, Wang, Zhang, Badie, Kalkum, He, Yi and Shi81) (Figure 2). Moreover, the demethylation of m1A by α-ketoglutarate-dependent dioxygenase human AlkB homolog 3 (ALKBH3) has been identified as crucial for the nucleosome formation of the promyelocytic leukemia (PML) protein, offering a novel therapeutic approach (Ref. Reference Gu, Zhuang, Yu, Yang, Ge, Ruan, Jia, Fan and Chai118). The METTL1/WDR4 complex, a key regulator of m7G modification, has been connected to various types of cancers and presents itself as a promising candidate for cancer therapy (Ref. Reference Han, Huang, Li, Chen, Zhu, Pan and Zhang119). Also, the identification of altered patterns of rRNA pseudouridylation in BC suggests the potential use of pseudouridyl-modified rRNA sites for developing therapeutic strategies targeting BC (Ref. Reference Barozzi, Zacchini, Corradini, Morara, Serra, De Sanctis, Bertorelli, Dassi and Montanaro120). Moreover, studies indicate that the high expression of YBX1 in OC cells can recognize the m5C modification on CHD3 mRNA, and YBX1 inhibitor SU056 can reverse the platinum resistance in animal models, suggesting that inhibition of YBX1 may be a potential strategy to overcome platinum resistance in OC (Ref. Reference Meng, Miao, Zhang, Chen, Yuan, Wan, Jiang, Zhang and Cheng121). Lastly, the miRNA mimics demonstrate significant potential in the clinical trials for cancer therapy; however, they encounter challenges related to stability and off-target effects. The new generation of molecular mimics enhances the stability of RNA oligonucleotides and minimizes the off-target effects, thereby facilitating their clinical application (Ref. Reference Toden, Zumwalt and Goel122).

Figure 2. Regulatory agents in chemically modified RNAs. (A) The METTL3 inhibitor STM2457 inhibits cell growth, invasiveness, migration, and enhances cell apoptosis in intrahepatic cholangiocarcinoma (ICC). The FTO inhibitors CS1 and CS2 attenuate leukemia stem/initiating cell growth, self-renewal and immune evasion in multiple types of cancers. Combination of PD-L1 blockade and the FTO inhibitor Dac51 inhibits cell growth in in melanoma and lung cancer. The IGF2BP2 inhibitor JX5 suppresses the expansion of T-cell acute lymphoblastic leukemia (T-ALL). (B) HOXC8 activates NAT10 and induces the ac4C modification of FOXP1 mRNA, thereby enhancing the immunosuppressive properties of tumor-infiltrating regulatory T cells (Tregs). NAT10 knockdown contributes to the effectiveness of PD-L1 blockade efficacy, thereby suppressing cervical cancer (CC) progression. (C) The PUS7 inhibitor C17 inhibits cell growth and tumor progression in glioblastoma.

Chemically modified ncRNAs present significant potential as therapeutic targets; however, several practical challenges hinder the development of therapies based on these modifications. One promising strategy is exosome-mediated delivery, which leverages the natural capacity of exosomes to transport miRNAs and evade phagocytosis. Nonetheless, issues such as the immunogenicity associated with allogeneic exosomes, along with challenges related to large-scale production and high manufacturing costs, must be addressed. Furthermore, the pharmacokinetic properties of ncRNAs pose limitations on their clinical application. Although chemical modifications can enhance RNA stability and bioavailability, further optimization of their metabolism and distribution in vivo is necessary (Ref. Reference Hueso, Mallén, Suñé-Pou, Aran, Suñé-Negre and Navarro123). Additionally, synthetic nanoparticles, including lipid nanoparticles, have demonstrated potential in the delivery of ncRNAs; however, their effectiveness in targeting specific cells while minimizing off-target effects remains a concern (Refs Reference Yang, Liang and Tong124, Reference Barber, Pater, Gagnon, Damha and O’Reilly125). In terms of stability, chemical modifications, such as Nm modification, have proven effective in enhancing the stability of ncRNAs and reducing their immunogenicity (Ref. Reference Toden, Zumwalt and Goel122). Nevertheless, additional research is required to optimize these modifications and ensure their safety and efficacy in clinical applications (Ref. Reference Hueso, Mallén, Suñé-Pou, Aran, Suñé-Negre and Navarro123). Regarding potential side effects, the off-target effects of ncRNAs and their capacity to activate the immune system are significant issues (Ref. Reference Hossam Abdelmonem, Kamal, Wardy, Ragheb, Hanna, Elsharkawy and Abdelnaser126). Although advancements have been made in mitigating immunogenicity through chemical modifications, the long-term implications of these modifications on the immune system are not yet fully understood.