Functional regulation of GlycoRNA and progress in malignant tumors

Abstract

Background

Glycans form unique GlycoRNAs through covalent modification of RNA molecules. GlycoRNAs serve as key components in intercellular communication. Early diagnosis and treatment of malignant tumors have always been the focus of cancer research, and the emergence of GlycoRNA has brought new opportunities to this field.

Main body

With breakthroughs in detection technologies, the structure and function of GlycoRNA are being gradually revealed. Here we describe the multiple biological functions of GlycoRNA and discuss the newly discovered GlycoRNA present in the exosomal lumen. This article primarily focuses on the structure and functional regulation of GlycoRNA, as well as detection technologies, to illustrate its potential in tumor diagnosis and treatment.

Conclusion

Our aim is to provide new perspectives for GlycoRNA biology research and promote its applications in malignant tumors.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-07694-1.

Preface

Glycosylation is a common post-translational modification in living organisms. This dynamic process requires the involvement of multiple glycosylation pathways [1]. During glycosylation, various types of glycans are generated, and their structural diversity reflects the diversity of biological functions [2]. These structurally distinct glycans mediate a wide range of key biological processes, including the regulation of cell adhesion, molecular transport and clearance, signal transduction, and tumor immune surveillance, thereby ensuring the fine-tuned regulation of vital activities [3, 4].

In oncology, aberrant glycosylation is a hallmark of cancer cells. The process involves the formation of glycans, glycoproteins, and the regulation of glycosyltransferases. Such alterations can enhance the motility of cancer cells, modulate adhesion, and ultimately drive tissue invasion and metastasis [5, 6]. At present, most clinical tumor markers are associated with glycoproteins or glycans [7]. Early and accurate identification of these tumor-specific glycosylation alterations contributes to early diagnosis and therapeutic intervention, and helps prolong patient survival [8]. In addition, abnormal glycosylation structures can specifically bind to immune checkpoints, allowing tumor cells to evade immune surveillance and clearance [9].

Traditionally, glycans were thought to primarily modify lipids and proteins, while RNA was not considered to undergo glycosylation. However, Ryan Flynn and colleagues first discovered through sucrose gradient centrifugation that RNA can also serve as a carrier of glycosylation modifications in mammals. These glycan-modified RNAs are mainly distributed among small non-coding RNAs and are termed GlycoRNAs [10]. The discovery of GlycoRNA represents a significant breakthrough at the intersection of RNA biology and glycobiology. Although functional studies of GlycoRNAs remain at an early stage, there is already clear evidence that they play roles in immune regulation and inflammatory responses [11, 12]. In oncology, GlycoRNA participates in multiple mechanisms that regulate tumor proliferation and metastasis, highlighting its great potential in tumor diagnosis and therapy [13]. This review mainly discusses the structure and function of GlycoRNAs, including advances in detection technologies. It also summarizes their roles in tumor initiation and progression, as well as their potential applications in cancer treatment. Our goal is to provide new perspectives for future GlycoRNA research and to promote their translation into clinical practice.

Regulation of the structure and function of GlycoRNA

Structure of GlycoRNA

GlycoRNA has recently emerged as an important discovery in glycobiology, as glycosylation endows RNA molecules with new functional properties. GlycoRNA modifications are widely present across multiple types of small RNAs, with Y RNA being the most abundant subtype, while tRNA and snoRNA are considered among the most representative glycosylated RNA species [14, 15]. However, with the emergence of detection techniques such as Clier-seq, the GlycoRNA family has expanded to a broader range of RNA types, including vtRNAs and lncRNAs [16]. These diverse GlycoRNAs may localize to different cellular organelles [17]. In classical glycobiology, N-glycosylation and O-glycosylation are the best-characterized types of modifications [18]. N-glycosylation, in particular, has been extensively studied, with its synthesis initiated in the endoplasmic reticulum and further processed in the Golgi apparatus [19]. To investigate the composition of GlycoRNA, Ryan Flynn and his colleagues treated purified GlycoRNA with PNGase F and found that the attached glycans were chemically very similar to the N-linked glycans of glycoproteins [10]. These N-glycans are composed of various monosaccharides, with sialylation and fucosylation being the most common forms. Further analysis revealed that their glycans typically contain terminal sialic acid residues, which can be efficiently removed by sialidases [15]. This feature is consistent with N-linked glycans, suggesting that the glycans carried by GlycoRNA share similarities with those of glycoproteins in terms of both structure and biosynthetic pathways [20]. However, GlycoRNA is resistant to most commonly used glycosidases, which are typically able to efficiently remove N- or O-linked glycans from proteins [15]. Therefore, although the glycans carried by GlycoRNA are structurally similar to N-linked glycans, their covalent attachment to RNA does not follow the classical modes of protein glycosylation. GlycoRNA may utilize a unique chemical or enzymatic linkage mechanism [10, 21]. To further elucidate the mechanism of RNA–glycan linkage, researchers developed an optimized reaction-based probe labeling method, termed rPAL, using periodate oxidation and aldehyde-reactive chemistry. They discovered that the covalent attachment of N-glycans to RNA occurs through the modified nucleoside acp³U [22]. Meanwhile, enzymatic digestion experiments suggested that non-enzymatic covalent coupling between glycans and nucleic acids may also occur [15, 23]. However, the current evidence more strongly supports the notion that GlycoRNAs are primarily generated through metabolism-dependent enzymatic pathways.

O-glycosylation is another important post-translational modification involved in multiple cellular processes, including transcription, translation, and receptor recognition [24, 25]. Li and colleagues proposed new insights into the role of O-glycosylation in GlycoRNA composition. Using a combination of SPCgRNA and TnORNA techniques, they demonstrated that N- and O-glycosylation can coexist on the same miRNA, and a strong positive correlation was observed between FCN-GlycoRNA and FCO-glycoRNA [13]. In another preprint study, Jennifer Porat and colleagues identified O-linked glycans as major structural components of GlycoRNAs. They further showed that the O-glycosylation biosynthetic pathway plays a critical role in regulating the abundance of sialylated GlycoRNAs. Additional research revealed that O-glycosylated RNAs exhibit markedly increased sialylation in colonic organoids derived from patients with ulcerative colitis, suggesting that aberrant O-glycosylated RNAs are closely associated with disease states. Together, accumulating evidence indicates that both N- and O-glycans are essential components of GlycoRNAs. However, the molecular mechanisms underlying their interplay remain unknown, and glycosylation may involve more complex relationships that require further investigation.

Biosynthesis and trafficking of GlycoRNA

GlycoRNAs are primarily localized on the cell surface, with the highest abundance at the plasma membrane, although small amounts are also present in the luminal compartments of membranous organelles [10]. To further investigate their intracellular transport mechanisms, Ma and colleagues applied the ARPLA technique. They provided the first evidence that GlycoRNAs are autonomously generated within cells and transported to the cell surface through SNARE protein–mediated secretory exocytosis, rather than via intercellular transfer [26].

The biosynthesis and trafficking of GlycoRNAs is a complex and highly coordinated process, tightly regulated by multiple factors that also determine their intracellular stability and biological functions. Specifically, intracellular transport is regulated by several factors, with SIDT family genes playing central roles in RNA binding, uptake, and transmembrane transport [26–28]. When SIDT1 or SIDT2 is knocked down, the cell-surface GlycoRNA signal disappears, indicating that SIDT proteins are essential transmembrane RNA transport components required for GlycoRNA biogenesis [12]. Existing evidence supports that the production of GlycoRNA depends on several key enzymes involved in the canonical N-glycosylation pathway. For example, treatment with the OST inhibitor NGI-1 significantly reduces GlycoRNA signals, indicating that OST participates in the glycan transfer process of GlycoRNA. Similarly, inhibition of α-mannosidase I or II leads to decreased GlycoRNA signals and altered migration patterns [10]. Moreover, classical N-glycosylation pathways in the endoplasmic reticulum and Golgi apparatus also play essential roles in GlycoRNA synthesis. Several glycosyltransferases, including MGAT5B and B4GALT1, are responsible for glycan extension and modification. Loss of their function not only alters N-glycan structures but also indirectly affects GlycoRNA levels [29]. Acp³U has been confirmed as a key component of GlycoRNA. DTWD2 is the essential enzyme that installs the acp³U modification, and when acp³U levels decrease, both the molecular weight and the detection signal of GlycoRNA are simultaneously reduced [22]. However, when GlycoRNA-S was labeled using GalNAz, its signal was not affected by NGI-1 or kifunensine, indicating that it does not depend on acp³U modification [11]. In addition, key factors such as OST, N-glycan processing enzymes, DTWD2, and SIDT play roles in GlycoRNA glycan assembly, substrate modification, and transmembrane transport. Nevertheless, it remains largely unknown whether different types of GlycoRNA utilize the same enzymatic pathway and how substrate RNAs are selected. Increasing evidence supports that GlycoRNA formation relies on an enzymatic reaction system highly similar to classical protein N-glycosylation, including DTWD2, OST, and a series of N-glycan processing enzymes. GlycoRNA is highly sensitive to glycosylation inhibitors such as NGI-1, kifunensine, and tunicamycin, and is significantly affected by perturbations of genes including STT3A, MOGS, MGAT1, and SIDT1/2, suggesting an enzymatic glycosylation reaction system [10, 22, 30]. However, the possibility of non-enzymatic reactions or the formation of pseudo-glycosylated adducts during sample processing cannot yet be completely excluded. Therefore, drawing on the substrate recognition mechanism of protein N-glycosylation, it is possible that glycosyltransferases recognize specific RNA sequences or secondary structure motifs. RNA-binding proteins may act as molecular chaperones, delivering specific RNAs to the ER lumen for glycosylation. Future studies could employ CRISPR-based gene knockout or screening strategies to identify essential glycosyltransferases, glycan processing enzymes, RNA chaperones, and transmembrane transport factors. Meanwhile, quantitative analysis of RNA-linked glycans combined with rigorously controlled chemical probes can distinguish genuine enzymatic GlycoRNA from potential non-enzymatic adducts, thereby providing a more precise understanding of the biochemical mechanisms of GlycoRNA.

Functional diversity of GlycoRNA

The functions of GlycoRNA are primarily determined by their glycans, while the RNA fragments may act as “scaffolds” to support glycan–receptor interactions. Studies have shown that GlycoRNA glycans exhibit highly tissue-specific distributions, with significant differences in abundance and structural types, thereby conferring distinct functions to GlycoRNAs [31]. For example, based on differences in glycan composition, GlycoRNAs can be classified into GlycoRNA-L and GlycoRNA-S. GlycoRNA-L is enriched in sialic acids, whereas GlycoRNA-S contains higher levels of N-acetylglucosamine (GlcNAc). GlycoRNA-L is widely expressed across most tissues and cell lines, while GlycoRNA-S is restricted to certain specific tissues and cell types [11]. The diversity of glycan structures enables GlycoRNAs to interact with multiple receptors, such as members of the Siglec family and P-selectin. Both GlycoRNA-L and GlycoRNA-S can directly bind to Siglec-5, thereby promoting interactions between human monocytes and endothelial cells. Due to differences in the composition of free oligosaccharides, Siglec-5 exhibits significantly higher binding affinity for GlycoRNA-L compared to GlycoRNA-S [11].

Beyond these findings, one of the most remarkable new functions of GlycoRNA is its role in mediating communication between cells and the extracellular environment. Specifically, RNA-binding proteins (RBPs) present on the surface of living cells can form functional complexes with GlycoRNAs, referred to as GlycoRNA-csRBP clusters [32]. These clusters provide binding sites for cell-penetrating peptides, facilitating their entry into cells via endocytosis or other mechanisms, and regulating cell–environment interactions [33].

The stable presence of GlycoRNAs on the cell surface provides the structural foundation for executing these functions. Evidence suggests that the RNA fragments of GlycoRNAs may be protected by unidentified cell-surface proteins, conferring resistance to RNase degradation and maintaining stability [12]. However, exogenous sialidases or extracellular RNases can remove surface GlycoRNAs and disrupt GlycoRNA–csRBP complexes. This, in turn, impairs the internalization of cell-penetrating peptides such as TAT and weakens interactions between cells and their environment [10, 32, 33]. GlycoRNAs possess regulatory potential that endows them with diverse biological functions, enabling the fine-tuned modulation of cell–environment interactions and laying the foundation for their roles in immune regulation.

Role of GlycoRNA in immune and inflammatory responses

Glycoproteins are closely associated with immune functions and are involved in regulating the development, differentiation, and activity of immune cells [34–36]. As a newly discovered class of glycosylated RNA molecules, GlycoRNAs are highly expressed in immune organs such as the spleen, thymus, bone marrow, and lymph nodes [11]. Ma and colleagues applied ARPLA technology to visualize immune cell differentiation and immune responses. They found that GlycoRNA levels decrease correspondingly as immune cells mature. The removal of GlycoRNAs is associated with reduced adhesion of immune cells to vascular endothelial cells [30]. Notably, GlycoRNAs also regulate neutrophil capture, rolling, and transendothelial migration. The RNA fragments of GlycoRNAs act as critical adhesion molecules, influencing the recruitment of neutrophils to sites of inflammation, thereby playing an important role in inflammatory responses [12] (Fig. 1). Based on this regulatory function, Zhang and colleagues developed a novel nanoparticle therapy, GlycoRNA-NP-siMT1. These nanoparticles incorporate GlycoRNAs from HL60 cell membranes as core components, endowing them with neutrophil-mimicking properties. This enables their targeted accumulation in the abdominal aortic aneurysm (AAA) lesion microenvironment. The results showed that GlycoRNA-NP-siMT1 competitively inhibited neutrophil infiltration, significantly reducing neutrophil numbers in AAA lesions and decreasing the formation of neutrophil extracellular traps (NETs) [37]. Furthermore, GlycoRNAs participate in intercellular communication and immune regulation through specific recognition by immune receptors. For example, the sialic acid-binding immunoglobulin-like lectin (Siglec) receptor system can specifically recognize GlycoRNAs [10]. The Siglec receptor family plays a central role in mediating cell recognition and transmembrane signaling [38–40]. Studies have shown that in human monocytes, GlycoRNAs bind to Siglec-5, promoting adhesion between monocytes and activated endothelial cells [11, 41]. This process enhances monocyte–endothelial interactions associated with inflammatory diseases, which is critical for monocyte migration from the bloodstream to inflamed tissues [42, 43]. Meanwhile, another recognition receptor, P-selectin (Selp), has been found to functionally interact with GlycoRNAs. Selp is a lectin with high binding specificity for sialylated glycans [44, 45]. Glycan fragments of GlycoRNAs can serve as novel ligands for Selp, specifically binding to it and thereby modulating neutrophil–endothelial interactions [46, 47].

Fig. 1.

GlycoRNA can be specifically recognized by the immune receptor Siglec, and its glycan portion can also act as a novel ligand for Selp, thereby mediating and regulating neutrophil–endothelial cell interactions; GlycoRNA associates with specific csRBPs to form GlycoRNA–csRBP clusters, which provide binding sites for exogenous molecules such as the cell-penetrating peptide TAT

During immune responses, glycosylation modifications can also bidirectionally regulate immune activation and suppression by modulating glycan expression, significantly affecting overall immune responses [48–50]. However, beyond glycoproteins, GlycoRNAs represent a novel glycosylation-mediated mechanism for immune regulation. Recent studies have shown that N-glycosylation of GlycoRNAs can mask the acp³U nucleoside, preventing recognition of RNA by endosomal receptors TLR3 and TLR7, thereby averting excessive innate immune responses. Through this mechanism, N-glycosylation protects RNA from inducing aberrant activation of efferocytes, facilitating non-inflammatory clearance of apoptotic cells [51]. Conversely, loss of the key synthetic enzyme DTWD2 prevents RNA glycosylation, leading to hyperactivation of the immune system. Activation of this potential immune pathway may be associated with the development of autoimmune diseases [23, 51].

GlycoRNA in exosomes

Exosomes, as a special type of extracellular vesicle, can mediate intercellular communication and are involved in processes such as antigen presentation [52, 53]. Multiple studies have confirmed the cell surface localization of GlycoRNA. In addition to surface-localized GlycoRNA, research has also reported functionally distinct GlycoRNAs localized within exosomes. Sharma, et al. first demonstrated that GlycoRNAs exist not only on cell membranes but also within the luminal space of exosomal vesicles [54] (Fig. 2).

Fig. 2.

On the outer surface of sEVs, GlycoRNA directly participates in the recognition and adhesion to target cells, enhancing the efficiency of sEV internalization by recipient cells; Within exosomes, GlycoRNA serves as an important carrier for intercellular communication

GlycoRNA can participate in cell-cell communication by serving as ligand-binding molecules for intercellular signaling. The discovery of exosomal GlycoRNA further supports its function as an intercellular communication vehicle [30]. Sharma’s study revealed that exosomes serve as carriers for GlycoRNA transmission between cells, and through exosomal internalization functions, they enable the exchange of genetic information between cells. This novel finding demonstrates the crucial regulatory role of glycosylation modifications in RNA sorting into exosomes [54].

Studies on glycan precursor metabolic pathways shared by GlycoRNA and glycoproteins indicate co-regulatory mechanisms between them. GalNAc, the core precursor molecule for glycosylation modifications, has been proven to directly regulate GlycoRNA synthesis as its biochemical precursor [55]. Notably, when protein glycosylation processes are blocked, the generation of exosomal GlycoRNA correspondingly decreases. This observation aligns with Flynn R.A. et al.‘s proposition that cell surface GlycoRNA depends on the classical N-glycan biosynthesis machinery of glycoproteins [10, 54]. These discoveries provide new perspectives on the regulatory relationship between GlycoRNA and glycoproteins, establishing an important foundation for developing exosome-based diagnostic and therapeutic technologies.

GlycoRNA detection methods

GlycoRNAs possess diverse and complex functions within cells. To characterize their structure and function, researchers have continuously optimized detection methods, thereby expanding their potential for clinical applications (Table 1). Initially, Ryan R.A. et al. employed an azide-labeled sialic acid precursor (Ac4ManNAz) and, through azide–alkyne click chemistry, first identified GlycoRNAs in highly purified RNA samples [10]. Building on this strategy, they further applied a biotin-labeling approach combined with streptavidin magnetic beads, achieving highly specific enrichment of GlycoRNAs and improving detection sensitivity and reliability [15]. To further analyze modification types on RNA bases, they developed the periodate oxidation and aldehyde-reactive probe labeling (rPAL) technique coupled with mass spectrometry (SWATH-MS) for quantitative analysis. This approach revealed glycosylation modifications on RNA, particularly acp3U modifications on tRNAs, providing direct evidence of chemical linkages between RNA and glycans. However, it is more suitable for detecting GlycoRNAs containing sialic acid residues [22]. Later, a preprint by Li, et al. proposed lectin-binding detection (LBD), a method that leverages the specific binding of lectins to glycosylation modifications on GlycoRNA for direct GlycoRNA detection. Relative to metabolic labeling and rPAL methods, LBD exhibits higher sensitivity toward GlycoRNA, a more straightforward experimental procedure, and the capability to detect diverse glycoforms of GlycoRNA. Notably, building on previous research, Ma, et al. developed a novel in situ visualization method for GlycoRNA named ARPLA. This revolutionary technology first achieved high-sensitivity, high-selectivity visualization of GlycoRNA at single-cell resolution. By integrating dual probe systems for glycan recognition and RNA in situ hybridization, it enabled the observation of GlycoRNA colocalization with lipid rafts at subcellular levels. This revealed GlycoRNA’s spatial distribution on cell surfaces, intracellular trafficking mechanisms, and potential disease-related functions. This provides an unprecedented tool for GlycoRNA biology research [30]. In the metabolic labeling workflow, the Clier-seq technique proposed by Zhu, et al. integrates metabolic labeling with click chemistry to achieve highly sensitive enrichment and sequencing analysis of sialylated GlycoRNA. It effectively covers a transcript range of 50–2000 nt, thereby successfully expanding the identification scope of GlycoRNA from conventional small RNAs to novel molecules such as long noncoding RNAs. It should be noted that Clier-seq has limited detection efficiency for short RNAs under 50 nt, highlighting the need to fully consider the detection range and potential biases of different techniques when interpreting GlycoRNA research data [16]. Additionally, Liu, et al. accomplished in situ visualization of RNA-specific sialylation on live cell membranes through a hierarchical encoding strategy (HieCo2) combined with hybridization chain reaction (HCR). This strategy employs dual encoding with SC and RC probes plus HCR signal amplification, enabling highly sensitive and specific GlycoRNA detection with quantitative analysis of N-glycosylation sites [56]. Zhao, et al. applied molecule differentiation encoding microscopy (MDEM) with orthogonal tandem repeat DNA identifiers (OTRDI) to GlycoRNA, achieving nanoscale visualization of densely distributed glycosylated RNAs on cell surfaces. They discovered 17% of U1 glycosylated RNA molecules cluster in cell surface nanoenvironments, providing evidence for GlycoRNA functional studies [57]. Despite the rapid advancement of detection techniques, significant methodological heterogeneity remains. Different methods vary in their ability to recognize specific glycan types, spatial resolution, and sample processing workflows. Consequently, reports on the cellular localization, abundance, and function of GlycoRNA show a certain degree of inconsistency. Therefore, it is essential to maintain a critical perspective on the limitations and potential biases of current GlycoRNA detection methods.

Table 1.

Advantages and disadvantages of GlycoRNA assays

Detection Method	Principle	Sensitivity And Specificity	Glycan Type Coverage	In Vivo Applicability	Major Limitations	References
Metabolic Labeling	Utilizes azide chemistry with azide-labeled sialic acid precursors to detect glycoRNA in highly purified RNA.	High sensitivity; capable of detecting differences at endogenous levels across different cell lines.High specificity; relies on the sialic acid biosynthesis pathway for specific detection.	Primarily detects sialic acid, indirectly reflecting the underlying glycan structures it is attached to.	Applicable to live cells and live mouse models.	Only labels newly synthesized sialic acid; efficiency depends on metabolic labeling and cell state; glycan modification on RNA over 200 nucleotides.	[10]
rPAL-SWATH-MS	Combines periodate oxidation, aldehyde-reactive tagging, and mass spectrometry quantification to detect modification types on RNA bases.	Extremely high sensitivity; detection signal is about 25-fold higher than the Ac4ManNAz method.High specificity; highly specific for sialic acid-containing glycans.	Primarily detects glycans containing cis-diol structures, highly specific for sialic acid.	Suitable for fixed cells or RNA extracted from tissues, not for real-time imaging in live cells.	Limited to detecting sialic acid-modified glycoRNA; requires cell lysis and RNA extraction, losing spatial information.	[22]
ARPLA	Dual-probe system combining glycan recognition and RNA in situ hybridization, with fluorescence imaging via rolling circle amplification.	High sensitivity; can detect low-abundance glycoRNA and achieve single-molecule visualization on the cell membrane.Extremely high specificity; requires simultaneous presence and spatial proximity of both glycan and RNA recognition.	Specifically detects Neu5Ac-modified RNA, covers multiple glycoRNA sequences.	Applicable to various fixed cell lines, enables cell surface and subcellular localization.	Can only detect glycoRNA with known sequences; weak signal for low-abundance targets.	[30]
HieCo2-HCR	Employs a hierarchical coding strategy combined with hybridization chain reaction, using SC and RC dual coding and HCR signal amplification to detect glycoRNA.	Extremely high sensitivity; enables single-molecule detection on live cell membranes, HCR signal amplification significantly enhances sensitivity.Extremely high specificity; based on dual recognition by sialic acid aptamer and RNA.	Specifically detects sialic acid-modified RNA.	Enables real-time imaging in live cells, limited to in vitro cell culture.	Relies on exogenous metabolic precursors; complex probe design; primarily provides qualitative spatial information, limited quantitative capability.	[56]
MDEM-OTRDI	Molecular differential encoding microscopy combined with orthogonal tandem repeat DNA identifiers for digital analysis of biomolecules.	Extremely high sensitivity; through RCA signal amplification, can detect single biomolecules and accurately count copy numbers in nano-space clusters.High specificity; achieves precise differentiation labeling and quantification via specific probes.	Can detect sialic acid-modified glycans, detection type determined by labeling probes.	Suitable for fixed cells, not applicable for live-cell imaging.	Complex workflow, cannot image dynamic processes in live cells in real time.	[57]
GlycanDIA	Combines DIA technology and higher-energy collisional dissociation mass spectrometry to analyze the glycan profiles of glycoRNA.	High sensitivity; can detect very low concentration glycans, nearly 100 types of N-glycans can be identified.High specificity, capable of distinguishing glycan composition and structural isomers.	Broad coverage of various modifications including sialylation, fucosylation, high-mannosylation, and low-abundance modifications.	In vitro analysis, applicable to cell lysates and tissue samples.	Limited ability to resolve glycan linkage sequences and specific attachment sites due to glycosidic bond cleavage; potential biases from glycan release and purification processes.	[31]
SPCgRNA	Solid-phase chemoenzymatic method using galactose oxidase to target N-glycosylated RNA, enriching via hydrazide magnetic beads while maintaining RNA integrity.	Moderate sensitivity; capable of enriching and identifying hundreds of N-glycan RNAs terminating with Gal/GalNAc.High specificity; specific for N-glycan RNAs containing Gal/GalNAc residues oxidizable by GAO.	Primarily covers complex N-glycans containing Gal/GalNAc residues that can be oxidized by GAO.	Only applicable to RNA extracted in vitro, not suitable for live cells or in vivo imaging.	Cannot determine glycosylation sites or the specific structures of N-glycans.	[17]
TnORNA	Solid-phase chemoenzymatic method for capturing O-glycosylated RNA containing the Tn antigen.	Moderate sensitivity; capable of enriching and identifying hundreds of O-glycoRNAs.High specificity; specific for O-glycan RNAs containing Gal/GalNAc not capped by sialic acid.	Detects Tn antigen and Core 1/Core 3 structures in O-glycans; can be extended to sialylated O-glycans via neuraminidase treatment.	Only applicable to RNA extracted in vitro, not suitable for live cells or real-time in vivo detection.	Glycosidase activity subject to competitive inhibition; oxidation step may damage RNA.	[13]
Clier-seq	Based on metabolic labeling and click chemistry for enrichment and sequencing analysis of sialylated glycoRNA.	Extremely high sensitivity; can detect at single-transcript level, covering transcripts from 50–2000 nt.Extremely high specificity; effective differentiation of true glycoRNA signals from non-specific background via Ac4ManNAz unlabeled control groups and Celier-qPCR validation.	Detects sialic acid-modified glycoRNA.	Live cells and various tissue samples.	Low detection efficiency for short RNA less than 50 nucleotides.	[16]

To perform comprehensive glycomics analysis of GlycoRNA, Xie, et al. developed an innovative glycomics method called GlycanDIA. This method significantly improves glycan identification sensitivity and quantification accuracy by combining DIA technology with HCD-MS/MS, enabling detection of RNA samples containing low-abundance N-glycans. Applying GlycanDIA, researchers identified over 200 N-glycans on mouse tissue RNAs, each exhibiting distinct abundance and distribution patterns [31]. Furthermore, Yang, et al. established a solid-phase chemoenzymatic method (SPCgRNA) for enriching and identifying N-glycoRNAs. They subsequently upgraded it to achieve high-quality, specific recognition of glycosylated RNA substrates while maintaining RNA integrity. This technique, applicable to diverse RNA types, demonstrates high specificity and sensitivity. It employs galactose oxidase (GAO) to specifically target N-glycosylated RNA, with resulting oxidized structures coupling to hydrazide magnetic beads for solid-phase N-glycoRNA capture [17, 58].

For O-glycan identification, researchers developed a solid-phase chemoenzymatic method specifically enriching and isolating O-glycosylated RNAs. This approach uses covalent immobilization on solid supports to capture and enrich O-glycoRNAs. It then selectively cleaves Tn antigen-containing O-glycoRNAs using the specific O-glycosidase GalNAcEXO, enabling efficient O-glycoRNA capture, enrichment, and identification [13]. Remarkably, combining SPCgRNA with TnORNA technology allows simultaneous identification of N-linked and O-linked GlycoRNAs within cells, revealing some RNA molecules bearing both modifications [13, 58].

Beyond compositional and functional characterization, researchers created the online tool PONglyRNA for predicting N-linked and O-linked glycosylation sites in RNA. By analyzing RNA sequence features, PONglyRNA provides crucial technical support for investigating GlycoRNA functions and mechanisms, offering novel perspectives for RNA modification research [13]. Although GlycoRNA detection technologies have achieved a series of breakthroughs, the field is still at an early stage. Future studies are needed to detect GlycoRNA in different subcellular locations to generate a more comprehensive and accurate GlycoRNA atlas.

Role of GlycoRNA in malignant tumors

Glycosylation has long been a focal point in studies of tumor pathophysiology, participating in various aspects of tumor proliferation, metastasis, and immune evasion [59, 60]. Aberrant glycosylation has been established as one of the hallmarks of cancer [61–63]. As an emerging entity bridging the RNA and glycosylation fields, GlycoRNA has shown potential in functional studies within malignant tumors. Recent studies indicate that GlycoRNA possesses potential as a tumor biomarker. However, the mechanisms by which it regulates tumor proliferation pathways, as well as its viability as an effective target for precision therapy, still require further investigation. Although the specific mechanisms by which GlycoRNAs contribute to tumor progression remain incompletely understood, in-depth investigation of their functions is expected to provide novel strategies. Such studies may offer new insights for early diagnosis and targeted interventions in cancer.

GlycoRNA as a novel regulatory factor in tumors

Tumor cells often exhibit abnormal glycosylation patterns, with increased sialylation and fucosylation being characteristic features [64, 65]. High sialylation can promote immune evasion, metastasis, and resistance to radiotherapy and chemotherapy [66, 67]. Notably, the trends of GlycoRNA alterations differ from those of conventional glycans. Two studies on breast cancer demonstrated that as tumor malignancy increases, cell-surface GlycoRNA levels gradually decrease, whereas overall glycan sialylation levels show an upward trend [30, 68]. This suggests that GlycoRNAs may possess functional significance distinct from traditional glycosylation patterns. Similar phenomena have also been observed in lung cancer and leukemia models. In normal alveolar epithelial cells and bone marrow-derived mesenchymal stem cells (MSCs), sialylated RNA levels are relatively high. In contrast, they are significantly reduced in lung adenocarcinoma cells A549 (RRID: CVCL_0023) and chronic myeloid leukemia cells K-562 (RRID: CVCL_0004) [68].

In pancreatic cancer, GlycoRNA exhibits more complex modification patterns, with a significant increase in differentially expressed GlycoRNAs. Both N-glycan and O-glycan modifications of miRNAs are altered. In cancer cells, N-glycosylated and O-glycosylated RNAs carrying the Tn antigen are enriched, and the number of miRNAs simultaneously carrying both N- and O-glycan modifications is markedly higher than in non-cancerous cells. This suggests that different forms of glycosylation may act synergistically to confer greater functional diversity to GlycoRNA [13]. Overall, GlycoRNA exhibits unique modification complexity and expression specificity in tumors. Leveraging this feature provides new avenues for its application in cancer diagnostics. Changes in glycosylation give rise to tumor-associated glycans or glycoproteins, which serve as important biomarkers for cancer detection and monitoring [69]. Common clinical tumor markers such as AFP, CA125, CEA, and PSA play a key role in early diagnosis and prognostic assessment [70, 71].The significant differential expression of GlycoRNA between tumor and normal tissues highlights its diagnostic value. Ren and colleagues developed a dual-recognition fluorescence resonance energy transfer (drFRET) technique to visualize GlycoRNA on small extracellular vesicles derived from serum. The results showed that GlycoRNA signals were markedly higher in patients with various cancers compared to non-cancer controls. Importantly, drFRET detection allows for pan-cancer screening using minimal amounts of bodily fluids. It can accurately distinguish six cancer types using a combination of multiple GlycoRNA features, demonstrating the potential of GlycoRNA-based liquid biopsy for early cancer diagnosis [72]. Considering the tumor-specific glycosylation patterns of GlycoRNA, dynamic monitoring of modifications such as sialylation can reflect therapeutic responses and disease recurrence, providing valuable guidance for treatment evaluation. In addition, GlycoRNA also participates in remodeling the tumor microenvironment by regulating the endocytosis of small extracellular vesicles (sEVs). As important mediators of cell–microenvironment communication, EVs carry a variety of bioactive molecules and can influence the functions of recipient cells by modulating signaling pathways and transcription [73, 74]. Studies have shown that GlycoRNA on the surface of sEVs is crucial for their internalization by recipient cells, and the removal of either glycans or RNA significantly impairs sEV uptake [72]. We speculate that GlycoRNA, by promoting efficient sEV internalization and their perinuclear localization, may influence nuclear signaling or metabolic states in recipient cells. This, in turn, could indirectly regulate the activation of tumor-associated fibroblasts (CAF) and associated stromal signaling pathways. Such regulation may mediate metabolic remodeling, epithelial–mesenchymal transition (EMT), extracellular matrix reconstruction, and the formation of pre-metastatic niches (PMN) [75, 76] (Fig. 3).

Fig. 3.

GlycoRNAs on the surface of sEVs mediate the internalization of recipient cells, contributing to PMN formation, tumor microenvironment remodeling, and distant metastasis

The tumor-specific regulatory functions of GlycoRNA

GlycoRNA plays a critical role in tumor proliferation, metastasis, and the remodeling of the tumor microenvironment. Evidence has shown that GlycoRNA exerts important functions in various malignancies, including pancreatic ductal adenocarcinoma (PDAC), hepatocellular carcinoma (HCC), colorectal cancer (CRC), and acute myeloid leukemia (AML). These studies reveal that GlycoRNA exhibits cancer type–specific functions across different tumor contexts.

Hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is an aggressive malignancy with a poor clinical prognosis. Therefore, it is of great significance to investigate the molecular mechanisms underlying its development and progression, and to explore novel therapeutic targets [77]. A recent study revealed that in HCC, cell surface RNAs, including GlycoRNA, are involved in driving tumor malignancy. Removal of surface RNAs significantly inhibited the proliferation and migration of liver cancer cells. Moreover, elimination of surface RNAs led to significant alterations in multiple key metabolic and signaling pathways within HCC cells, including apoptosis, signal transduction, ribosomal function, endocytosis, and autophagy-related pathways. Notably, amino sugar and nucleotide sugar metabolism, as well as taurine and hypotaurine metabolism pathways, were markedly dysregulated [78]. In addition, the study identified the RNA-binding protein SERBP1 as a key regulator of cell surface RNAs, including GlycoRNA. Knockdown of SERBP1 significantly reduced the migratory and proliferative abilities of HCC cells, further supporting the crucial role of surface GlycoRNA in the biological processes of HCC [78] (Fig. 4).

Fig. 4.

Upon removal of surface RNAs, multiple key metabolic and signaling pathways within hepatocellular carcinoma cells were significantly altered, leading to reduced proliferation and migration capabilities

Pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDAC) is the most common and aggressive subtype of pancreatic cancer, and its malignant progression is closely associated with abnormal molecular regulation [79]. Differential analysis of GlycoRNA in PDAC identified 12 significantly altered glycosylated miRNAs, whose target genes were enriched in key pancreatic cancer–related pathways, including the p53, FoxO, and JAK/STAT signaling pathways [17]. These pathways play critical roles in the progression of pancreatic cancer. In PDAC, the p53 pathway functions as a tumor suppressor by regulating the cell cycle, inducing apoptosis, and maintaining genomic stability; its mutation or inactivation promotes cancer progression and drug resistance [80, 81]. The FoxO pathway affects tumor cell survival by regulating cell proliferation, apoptosis, antioxidant capacity, and angiogenesis [82]. The JAK–STAT pathway promotes pancreatic cancer progression by enhancing cell proliferation and inhibiting apoptosis [83, 84]. These findings suggest that glycosylated miRNAs may promote tumor initiation, progression, and the development of drug resistance by regulating key signaling pathways [17]. In addition, researchers used a combined chemical and enzymatic TnORNA method to identify multiple miRNAs carrying Tn-O-glycosylation modifications. Target gene enrichment analysis showed that the targets of these glycosylated miRNAs were significantly associated with classical cancer-related signaling pathways such as PI3K–Akt and JAK–STAT. This suggests that they may be involved in regulating tumor cell proliferation and related processes, but the specific molecular mechanisms remain to be further elucidated [13] (Fig. 5). Moreover, the regulation of glycosylated miRNAs involves not only modifications of the RNA itself, but is also tightly controlled by glycosyltransferases [85]. For example, the glycosyltransferase B4GALT1 regulates the expression of glycosylated miRNAs, and affects cell cycle progression by inducing G2/M arrest, reducing the proportion of cells in the G1 phase, and promoting apoptosis [17]. In contrast, NGI-1, an inhibitor of oligosaccharyltransferase (OST), can increase apoptosis and suppress the cell cycle and cell proliferation [17]. The tumor-specific and differential expression of glycosylated miRNAs in pancreatic cancer, together with the enrichment of glyco-miRNAs in key signaling pathways, makes them promising therapeutic targets.

Fig. 5.

The glycosyltransferase B4GALT1 regulates the expression of glycosylated miRNAs, affecting cell cycle progression and promoting apoptosis in pancreatic cancer cells

Colorectal cancer

GlycoRNA profiling in colorectal cancer tissues revealed that both cancerous and adjacent normal tissues contain a wide variety of GlycoRNAs, with approximately 70% of glycosylated miRNAs shared between them. However, there are significant differences in the expression abundance of specific molecules. It was found that N-glycosylated hsa-miR-27a-3p plays a key regulatory role. This molecule interferes with the function of the circular RNA hsa_circ_0004194 through a competitive inhibition mechanism. Under normal conditions, hsa_circ_0004194 helps maintain stable expression of the nuclear receptor protein RXRα. Upon ligand activation, RXRα can suppress the transcriptional activity of the oncogene β-catenin, thereby inhibiting tumor growth. However, when N-glycosylated miR-27a-3p binds to hsa_circ_0004194, it results in a decrease in RXRα protein levels. This relieves the inhibitory effect of RXRα on β-catenin signaling, thereby activating the pathway and promoting the progression of colorectal cancer [86] (Fig. 6).

Fig. 6.

N-glycosylated miRNAs regulate the RXRα/β-linker signaling pathway, thereby driving the progression of colorectal cancer

Acute myeloid leukemia

The surface of cancer cells not only displays classical glycoprotein antigens but also presents complex structures composed of GlycoRNAs and cell surface csRBPs. In acute myeloid leukemia (AML), nucleophosmin (NPM1) is highly enriched in leukemic progenitor and stem cells. NPM1 forms specific nanoclusters with GlycoRNAs on the cell surface, which are absent in normal hematopoietic stem cells [87]. This suggests that GlycoRNAs may act as molecular scaffolds to stabilize the surface presentation of csRBPs. Targeting these complexes enables the specific recognition and elimination of tumor cells. Based on this finding, monoclonal antibodies have been developed that can selectively eliminate NPM1-positive tumor cells. In AML and NPM1-positive solid tumor models, this strategy significantly prolonged survival and reduced disease burden without damaging normal tissues [87]. Because NPM1 expression is not dependent on genetic mutations, GlycoRNA–RBP complexes may serve as universal therapeutic targets across various hematological malignancies and solid tumors [87]. Moreover, these GlycoRNA–csRBP clusters provide binding sites for cell-penetrating peptides. Precisely modulating GlycoRNAs could potentially influence cellular internalization processes, thereby regulating drug delivery efficiency [32, 33].

Challenges and future directions

Although recent innovations in GlycoRNA detection technologies have injected substantial momentum into this emerging field, significant limitations still exist in the current methodological toolbox. Metabolic labeling, rPAL, ARPLA, and HieCo2-HCR detection techniques rely on the specific recognition of Neu5Ac, potentially overlooking other glycoforms. Additionally, metabolic labeling can partially exclude nonspecific binding through rigorous controls, but signals from low-abundance targets remain susceptible to background interference, which may lead to false-positive or false-negative identification results.

Another controversial issue concerns the subcellular localization of GlycoRNA. While multiple studies have demonstrated the presence of GlycoRNA on the surface of live cells, the observed intracellular distribution, vesicular localization, and presentation within the endomembrane system vary significantly across studies (Fig. 2). Several studies show that GlycoRNA signals on live cell surfaces can be abolished by RNase or sialidase treatment, while cell viability remains unaffected, indicating that both the RNA and glycan portions are exposed to the extracellular environment. Methods such as lectin proximity labeling and Siglec recognition have successfully labeled GlycoRNA. These independent and methodologically diverse lines of evidence collectively support the localization of GlycoRNA on the outer leaflet of the plasma membrane.

Further investigation indicates that GlycoRNA is not uniformly distributed across the membrane surface but is localized within lipid rafts and can form stable nanoclusters with csRBPs. This provides an important structural basis for its function in extracellular signal regulation. Additionally, some studies report that GlycoRNA exists on the membrane surface of sEVs. Notably, certain GlycoRNA molecules are located within the lumen of exosomes, suggesting that GlycoRNA can exist not only on the cell surface but also inside vesicles, participating in intercellular communication. In contrast, the presence of GlycoRNA within the intracellular membrane system remains highly uncertain. GlycoRNA has been observed in association with transport-related SNARE proteins and enriched in compartments containing membrane structural components, suggesting potential transport and secretion via the intracellular membrane system. However, it is unclear whether these signals originate from GlycoRNA biosynthesis and transport pathways or from endocytic recycling. Future studies will require in situ high-resolution imaging, real-time tracking, and molecular labeling tools to elucidate the origin, transport routes, and functional differences of GlycoRNA in different subcellular localizations, ultimately establishing a model for GlycoRNA localization.

Many studies report that glycans can be covalently attached to RNA and propose a model in which GlycoRNA is covalently linked to the acp3U-modified base of RNA through classical N-glycosylation. However, some reports indicate that certain GlycoRNA-S are sensitive to O-glycosidase, suggesting the possible existence of multiple linkage types. The linkage mechanism itself still contains several key unresolved issues. For example, it is currently proposed that the carboxyl terminus of acp3U must first undergo amidation, forming a carboxamide structure similar to the asparagine side chain, in order to be recognized by OST and linked to the core GlcNAc. However, there is no direct evidence for this amidation step, and the enzyme responsible has not yet been identified.

More importantly, recent studies have raised strong doubts about the authenticity of GlycoRNA, bringing new controversy to the field. Kegel, et al. demonstrated that under denaturing conditions, protease digestion can completely abolish GlycoRNA signals, whereas under non-denaturing conditions, it cannot. This suggests that previously detected signals were likely derived from incompletely digested glycoprotein contaminants rather than genuine covalent glyco-RNA complexes. Additionally, proteomic analysis identified multiple typical glycoproteins, including LAMP1, LAMP2, and CD63, but failed to detect any specific GlycoRNA [88].

Conclusions and challenges

The discovery of GlycoRNA not only expands the molecular boundaries of glycobiology but also reveals a novel mode in which RNA, as a carrier of glycans, participates in cell surface signal transduction. Cell-surface GlycoRNAs are recognized by sialic acid-binding immunoglobulin-like lectin receptors, thereby regulating intercellular adhesion, signal transduction, and immune responses. Differences in glycan composition confer a high degree of functional specificity to GlycoRNAs. In recent years, the development of detection technologies such as ARPLA and SPCgRNA has enabled the visualization and high-sensitivity analysis of GlycoRNAs at the single-cell level, providing crucial tools for elucidating their biological functions.

Increasing evidence highlights the significant potential of GlycoRNAs in tumor diagnosis and therapy. In multiple tumor models, the expression levels and glycosylation patterns of GlycoRNAs are closely associated with tumor proliferation and metastasis. Although the precise relationship between GlycoRNA and tumors remains incompletely understood, new evidence suggests that complexes formed by GlycoRNA and csRBP may serve as specific molecular markers for immune clearance. These complexes may also function as molecular scaffolds or binding sites to regulate cellular signaling and drug delivery. In addition, based on their tumor-specific glycosylation patterns, dynamic monitoring of GlycoRNA modifications could provide a potential indicator for evaluating treatment response and efficacy.

Despite the promising prospects, this field still faces numerous challenges. Currently, detection methods for GlycoRNA require further optimization, and its specific mechanisms of action in tumors remain unclear. Therefore, a deeper understanding of the functions of specific GlycoRNA modifications in different tumor types and within the tumor microenvironment is crucial. Future research should further investigate the potential roles of GlycoRNA in immune evasion, intercellular communication, and targeted therapies. Integrating multi-omics approaches to map a comprehensive GlycoRNA atlas will help systematically elucidate its biological mechanisms and promote the identification of relevant therapeutic targets.

In summary, although GlycoRNA research is still in its early stages, it has already provided new insights into immune regulation and the pathophysiological mechanisms of cancer. In-depth analysis of its mechanisms of action is expected to advance GlycoRNA from basic research to clinical applications, laying the foundation for the development of novel diagnostic tools and therapeutic strategies.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contributions

ZL、XW、CM: Literature collation, article writing, and preparation of pictures; SZ: Prepare pictures and financial support; DG、YL: review manuscripts; YF: review manuscripts, financial support;

Funding

This study was supported by (1) Key project fund of Jiangsu Provincial Health Commission (ZD2022052, K2023016); (2) Suqian science and technology support project fund (KY202203); (3) Zhenjiang Key Research and Development Fund (SH2024002, SH2024075, JC2024031).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors have read and approved the final manuscript.

Competing interests

The authors declare no competing interests.