Advances in the study of nucleolar small RNAs in colorectal cancer

Summary

Colorectal cancer (CRC) is an important public health challenge worldwide, which has led to a continuous search for more precise diagnostic strategies. Recent studies have revealed that small nucleolar RNAs (snoRNAs) play an important role in CRC. In this article, we first summarize the molecular characteristics and physiological functions of snoRNAs, and then we discuss their mechanisms of action in colorectal carcinogenesis. A key focus is on the emerging potential of snoRNAs as liquid biopsy biomarkers and their prospects for clinical translation in cancer management.

Graphical abstract

Molecular biology; Cancer

Introduction

Small nucleolar RNAs (snoRNAs), first identified in Saccharomyces cerevisiae in 1991,¹ constitute a class of non-coding RNAs (ncRNAs) situated in nucleoli. They have piqued researchers’ interest in RNA epitranscriptomic regulation. Their evolutionary origins can be traced back to Archaea.² The typical length of snoRNAs is 60–300 nucleotides. SnoRNAs are divided into two main subclasses, namely box C/D - type snoRNAs and box H/ACA-type snoRNAs, according to their distinct structural characteristics, nucleotide compositions, and functional properties.³ Interestingly, a substantial number of snoRNAs, known as orphan snoRNAs, have no identified cellular function.⁴ Although they have complete box structural elements, their transcripts lack obvious ribosomal RNA (rRNA) modification targets, and the biological functions of these ncRNAs remain largely unknown. A great deal of research indicates that snoRNAs are indispensable for the post-transcriptional processing of ribosomal RNAs and small nuclear RNAs (snRNAs). They perform vital functions in maintaining RNA stability and enabling proper functional operation.⁵

Recent research has revealed that certain snoRNAs show differential expression in clinical samples and cell lines from a range of cancers (such as hepatocellular carcinoma (HCC) and breast cancer (BC)).^6,7 This finding underscores the potential of snoRNAs as prognostic and diagnostic markers while suggesting their crucial role in transcriptional and epigenetic regulation. Their altered function and expression are intimately linked to many illnesses, with substantial evidence indicating critical regulatory functions in the onset and progression of diverse malignancies.⁸ As a result, the aberrant expression of certain snoRNAs in blood is hypothesized to serve as diagnostic markers.

Colorectal cancer (CRC), a significant global health burden, arises from a multifactorial pathogenesis involving genetic susceptibility, lifestyle influences, gut microbial dysbiosis, and chronic inflammatory states.^{9,10,11,12,13,14} Its early clinical presentation is often marked by nonspecific symptoms leading to frequent late-stage diagnosis, and despite a combination of available treatments, including surgery and systemic therapies, therapeutic efficacy remains limited by primary resistance, disease recurrence, and metastasis.^15,16,17

The five-year survival rate for early-stage CRC is approximately 90%, falling sharply to 14% for advanced disease and underscoring the critical need for early detection.¹⁸ However, existing screening approaches—including fecal occult blood tests, colonoscopy, and liquid biopsies—remain limited by invasiveness, suboptimal sensitivity, or inadequate specificity.^19,20,21,22

Therefore, the discovery of efficient diagnostic markers for CRC is crucial to enable early detection of the disease, thereby reducing mortality. Preliminary findings indicate that the methylation profiles of circulating free DNA and circulating tumor DNA, including Myosin-Ig (MYO1G) and NDRG family member 4 (NDRG4) genes, hold promise for early-stage diagnosis, while the methylation status of TMEFF2, COL6A2, ZNF671, MAL, and AKR1B1 may improve detection sensitivity and accuracy.^23,24,25 Research into microRNAs (miRNAs) and circular RNAs has advanced significantly, with circ_0011536 showing an Area Under the Curve (AUC) of 0.982, and the plasma membrane-associated A2 protein combined with carcinoembryonic antigen (CEA) potentially outperforming conventional biomarker panels.^26,27 Recent studies also associate snoRNAs with CRC development, linking their aberrant expression to tumorigenesis and offering new insights into pathogenesis and therapy.²⁸ This study will examine the structural features, functional roles, and molecular mechanisms of snoRNAs in CRC, exploring their diagnostic and therapeutic applications to provide new perspectives for early diagnosis and clinical management.

Structure and function of snornas

Structural features of small nucleolar RNAs

Based on their distinct nucleotide motifs and secondary structural configurations, snoRNAs are primarily categorized into two classes: C/D box snoRNAs, denoted as SNORD, and H/ACA box snoRNAs, designated as SNORA.³

C/D box snoRNA lengths typically range between 60 and 120 nucleotides, containing two highly conserved elements: the C box (5′-RUGAUGA-3′, where R represents A or G) and the D box (5′-CUGA-3′). These two elements can form a stem-loop-stem structure through base pairing, known as the “kink-turn” or “K-turn” motif.^29,30

Most members also contain lower-conservation C′ and D′ boxes, flanked by the core elements—the C′ box resides 10–20 nt downstream of the C box, and the D′ box lies immediately upstream of the D box.³¹ The structural integrity of the snoRNA is conferred by its assembly with four core proteins: the methyltransferase fibrillarin (FBL), nucleolar protein 56 (NOP56), nucleolar protein 58 (NOP58), and the Small Nuclear Ribonucleoprotein 13 (SNU13), forming a stable snoRNP complex.^32,33

The vast majority function as antisense snoRNAs, featuring a >21 nt antisense element upstream of the C/D box that base-pairs with complementary target RNAs to guide their 2′-O-methylation (Nm) (Figure 1).³⁴

Figure 1.

Structure of C/D box snoRNA and C/D box snoRNP

(A) Structure of C/D box snoRNA.

(B) Structure of C/D box snoRNP.

H/ACA box-containing snoRNAs exhibit a characteristic molecular architecture defined by two primary structural elements: a conserved nucleotide length and a distinct conformational organization. These non-coding transcripts typically range in length from 120 to 140 nucleotides, with their functional sensitivity determined by a unique double-hairpin-hinge-tail configuration. The structural blueprint comprises: (1) a central single-stranded hinge region containing the H box (ANANNA) motif, which serves as a critical recognition site for protein partners, and (2) a terminal ACA triplet positioned three nucleotides upstream of the 3′ end.³⁵ This structure allows the H/ACA box snoRNA to bind to the core protein, forming the snoRNP complex. Distinct from C/D box small nucleolar ribonucleoprotein complexes (snoRNPs), H/ACA box snoRNAs are characterized by a unique protein cohort comprising dyskerin (a pseudouridine synthase also termed congenital dyskeratosis protein), glycine-arginine-rich protein 1, non-histone chromosomal protein 2, and nucleolar protein 10 (NOP10).^36,37,38 The H/ACA box snoRNAs bind to the target RNAs through their alternatively spliced exons and direct the conversion of uridine to pseudouridine, resulting in pseudouridylating rRNA (Figure 2).³⁰

Figure 2.

Structure of H/ACA box snoRNA and H/ACA box snoRNP

(A) Structure of H/ACA box snoRNA.

(B) Structure of H/ACA box snoRNP.

In addition to classical snoRNAs, researchers have identified specialized snoRNA isoforms, such as Cajal body-specific RNAs (scaRNAs). These are situated inside the intranuclear Cajal bodies and possess specific C/D box and H/ACA box sequence components, which play critical roles in snRNA modification and post-transcriptional regulatory processes.³⁹ Although there have been significant advancements in the research of scaRNAs, numerous questions remain unresolved, such as the sensitivity of subcellular localization and the mechanism of interaction with other intranuclear vesicles, which still need to be explored in depth.

Genomic structure of small nucleolar RNAs

The genomic structure of snoRNAs shows significant species-specificity. It has been shown that the majority of snoRNA genes in Saccharomyces cerevisiae exist independently as single cis-transcripts, only 20% are transcribed as multiple cis-transcripts, and only 11% are located in intronic regions.⁴⁰ In contrast, around 90% of human snoRNA genes are situated in the intronic regions of protein-coding genes or long ncRNAs.

SnoRNA maturation occurs through two distinct pathways: splicing-dependent and splicing-independent.⁴¹ In the splicing-dependent pathway, introns in precursor mRNAs are spliced out to form lariat structures, which are then linearized by the debranching enzyme to release the snoRNA-containing intronic RNA.⁴² The snoRNA is subsequently trimmed at its ends by exonucleases and assembled with core proteins—such as FBL, NOP56, NOP58, and SNU13 for C/D box snoRNAs, or dyskerin, GAR1, NHP2, and NOP10 for H/ACA box snoRNAs—to form functional snoRNPs.⁴³ In contrast, the splicing-independent pathway involves the cleavage of snoRNAs directly from larger transcriptional precursors, facilitated by characteristic stem-loop structures at their processing sites.⁴¹

Of interest, a few snoRNA genes, such as U3, U8, and U13 (including telomerase-associated snoRNAs) have independent promoters.⁴⁴

Multiple biological functions of small nucleolar RNAs

Beyond their well-established roles in Nm and pseudouridylation (Ψ), recent research has revealed that snoRNAs possess a spectrum of additional biological activities (Figure 3).

Figure 3.

The function of snoRNAs. (Created in BioRender)

(A) Classical snoRNA Functions.

(B) Non-Classical snoRNA Functions.

(C) Function of snoRNA-Derived small RNAs.

N4-acetylcytidine modification

The N4-acetylcytidine (ac4C) modification is an evolutionarily conserved RNA modification found in rRNAs, transfer RNAs (tRNAs), and messenger RNAs (mRNAs), where it facilitates ribosome biogenesis and modulates translational control.³ Catalyzed by the conserved NAT10/Kre33 acetyltransferase family, this modification is site-specifically installed at defined positions in 18S rRNA (e.g., C1842 in humans; C1773 in yeast) through the guidance of box C/D snoRNAs, which include U13 in mammals and snR4/snR45 in yeast.^45,46,47,48 Functional validation in HCT116 cells demonstrates that U13 depletion diminishes 18S rRNA acetylation, underscoring the essential role of snoRNAs in directing ac4C incorporation.⁴⁴ Beyond its established role in RNA modification, NAT10 also contributes to mitotic progression and DNA damage repair (DDR), revealing the functional versatility of this conserved molecular system.^45,46

Regulating selective splicing of messenger RNAs

In gene expression regulation, snoRNAs are involved not only in rRNA modification but also in the selective splicing of mRNAs, thereby regulating gene expression. Ogren et aldemonstrated that the expression level of SNORD94 directly affects the methylation level of the C62 locus on the U6 RNA and regulates mRNA variable splicing.⁴⁹ In addition, Kishore et al. found that SNORD115 was able to regulate the selective splicing of 5-hydroxytryptamine receptor 2C mRNA by competing with ADARase for binding sites.⁵⁰ Similarly, the study by Falaleeva et al. demonstrated the ability of certain snoRNAs to influence mRNA splicing; for example, SNORD27 influenced the splicing pattern of E2F7 mRNA by directly interacting with the E2F7 precursor mRNA and repressing the inclusion of specific exons.⁵¹

Oxidative stress

Emerging evidence reveals that specific snoRNAs function as key regulators in the oxidative stress (OS) response network. Under palmitate- or hydrogen peroxide-induced OS, snoRNAs, including U32a, U33, and U35a, are markedly upregulated, and their depletion enhances cellular resistance to lipotoxicity, potentially through the modulation of lipid stress responses linked to DDR.⁵² Further studies showed that doxorubicin-induced OS triggers the rapid cytoplasmic accumulation of ribosomal protein L13a (Rpl13a) -derived snoRNAs (U32a, U33, U35a), mediated by nuclear-derived superoxide and NADPH oxidase splice variants.⁵³ Thus, a snoRNA-mediated OS response pathway emerged, spanning from “nuclear transcription induction” to “cytoplasmic functional execution.”

Ribonuclease T2 also fine-tunes this network,⁵⁴ while in vivo knockout of the Rpl13a snoRNA cluster was shown to reduce reactive oxygen species and strengthen antioxidant defense.⁵⁵ Other snoRNAs, such as SNORA73 and SNORD46, are involved in OS regulation by different mechanisms, reflecting the multifaceted roles of snoRNAs in OS adaptation.^55,56,57

DNA damage repair

The network of DDR pathways is essential for maintaining genome stability, as it identifies and repairs a wide array of DNA lesions. DNA damage manifests in several forms, with single-strand breaks (SSBs) and double-strand breaks (DSBs) as the two principal types, each addressed by separate repair processes.⁵⁸ While SSBs arise more commonly, they are primarily corrected by mechanisms that rely on poly ADP-ribose polymerase (PARP).^59,60,61,62 DSBs, on the other hand, present a far more serious risk to genomic integrity. These critical breaks are mended through two key mechanisms: homologous recombination and non-homologous end joining. The accuracy of these repairs is overseen by core regulators such as ATM, DNA-PKcs, and RAD51.^63,64

SnoRNAs contribute to the DNA damage response (DDR) through various regulatory mechanisms. Regarding PARP-1, SNORA37/73A binds its DNA-binding domain to potentiate PARylation activity,⁶⁵ SNORA64 facilitates auto-PARylation-driven activation,⁶⁶ and SNORD104 enhances transcript stability via Nm guidance.⁶⁷ Within the PIKK kinase family, U3 snoRNA guides the phosphorylation of DNA-PKcs at a specific site, Thr2609, while SCARNA 2 impedes the assembly of the DNA-PK complex and simultaneously modulates ATM/ATR signaling.^68,69 Together, these results uncover a previously unknown circuitry of snoRNAs within the DDR, offering a fresh perspective on their contribution to genomic integrity.

P53 acts as a highly versatile nuclear transcription factor, central to preserving cellular homeostasis in response to diverse stressors, including DNA damage, OS, and metabolic imbalance. Although direct evidence linking snoRNAs to the p53-mediated DNA damage response remains sparse, recent work suggests they may influence p53 stability through post-translational modifications. Documented instances include SNORD50 A/B promoting TRIM21-mediated ubiquitination,⁷⁰ SNORA18L5 disrupting the p53-MDM2 interaction by sequestering RPL5/1171, and SNORA24 facilitating proteasomal degradation.⁷¹ Furthermore, the snoRNA-derived sno-MiR-28 adds a further layer of regulation by directly targeting TAF9B.⁷²

Together, these observations delineate an extended snoRNA-guided regulatory network that fine-tunes DDR beyond canonical repair pathways, providing new perspectives on the epigenetic control of genomic integrity.

Cell cycle and cell death

SnoRNAs play a crucial role in regulating cell cycle progression by modulating the expression levels of key regulatory complexes, such as cyclin D-CDK4/6 and cyclin E/A-CDK2, during the G1/S phase transition. Research has shown that specific snoRNAs, including SNORA18L5,⁷³ SNORD114,⁷⁴ SNORD76,⁷⁵ and SNORA21,⁷⁶ regulate key factors at the G1/S checkpoint. This regulation is achieved through their ability to modulate p53 phosphorylation, Rb phosphorylation levels, and CKI expression, which, in turn, influence cell cycle progression.

The cyclin B1/CDK1 complex is a central regulator of the G2/M checkpoint. Research by Xu et al. revealed that SNORD47 triggers G2 phase arrest in glioblastoma by suppressing the cyclin B1/CDC25C/CDK1 axis.⁷⁷ In contrast, Li et al. reported that SNORD52 facilitates the G2/M transition in HCC by binding to CDK1 and stabilizing it through altered Thr161 phosphorylation and ubiquitination.⁷⁸ A separate mechanism in HCC was described by Zhu et al., who found that SNORA14A drives cell-cycle arrest and apoptosis by upregulating SDHB expression, thereby reducing intracellular succinate levels.⁷⁹ Furthermore, SNORD50A exerts an anticancer effect in BC by inducing an M-phase block by inhibiting key mitotic genes, including SMC5, ATRX, CENPE, and CENPF.⁸⁰ Collectively, these studies illuminate diverse mechanisms by which snoRNAs govern cell cycle progression, with a pronounced impact on the G2/M checkpoint. Despite the limited number of studies in this area, snoRNA regulation holds promise as a novel therapeutic target for cancer and other diseases.

Notably, some snoRNAs have been found to modify tRNAs,⁸¹ regulate chromatin remodeling,^82,83,84,85 apoptosis,^86,87,88 and participate in autophagy^56,89,90 and iron death processes.^91,92,93

In summary, snoRNA biological roles are multidimensional, spanning multiple aspects from RNA modification to cell fate determination, and elucidating their molecular mechanisms offers innovative targets for the development of precision therapeutic strategies for diseases.

Protein secretion

The interaction of ncRNAs with proteins is an important molecular mechanism for their functions, and snoRNAs are no exception. Traditionally, snoRNAs are mainly involved in rRNA modification; however, recent studies have revealed that they have a novel function in the regulation of protein secretion. The snoKARR-seq technology developed by He Chuan and Tao Pan’s team innovatively combines chemical cross-linking technology with an antisense oligonucleotide enrichment strategy to achieve highly efficient capture of snoRNA target RNAs, which breaks through the limitations of traditional research methods.⁹⁴

In this study, researchers dissected the dual molecular mechanism of snoRNA regulation of protein secretion. The results show that SNORA73 recognizes target mRNAs with MBM motifs and forms a ternary complex with 7SL RNA via the 7BM site, thereby enhancing the interaction of mRNAs with signal recognition particles and promoting secretory protein transport. This study breaks new ground and reveals a new role for snoRNAs, which may provide a framework for interpreting the pathology of diseases related to secretory dysfunction or serve as a target for therapeutic strategies. However, the discovery of this mechanism also raises several scientific questions that are worthy of in-depth investigation. For example, whether the dual-binding mechanism is universal and whether there are differences in regulation under different conditions.

Generation of small nucleolar RNAs-derived microRNAs

In addition, some snoRNAs are subsequently processed to form stable short fragments, which are called snoRNA-derived microRNAs (sdRNAs).^95,96 Just like miRNAs, sdRNAs exhibit the ability to silence genes. They achieve this by identifying mRNA targets with the help of the RNA-induced silencing complex. As a result, they either repress translation or expedite the degradation of targeted mRNAs.⁹⁶ Emerging research indicates that sdRNAs can modulate oncogenic processes through various mechanistic pathways.⁹⁷

Development of small nucleolar RNAs in colorectal cancer

CRC ranks among the most prevalent malignant tumors globally. However, its stealthy initiation and the presence of non-specific symptoms during the early stages frequently cause it to be misdiagnosed as a benign intestinal disorder. This delay in diagnosis is a factor that leads to the relatively low five-year survival rate seen among patients with CRC, even though there has been significant progress and achievements in the areas of diagnostic and treatment methods.¹⁵ Consequently, it is of great urgency to develop new biomarkers to enable the early identification of diseases and related conditions. Numerous studies have shown that snoRNAs play a dual regulatory role in CRC development (Figure 4 and Table 1). In addition, due to their high stability in body fluids, including plasma, serum, and urine, and their relatively easy detection, these substances are well-suited to serve as circulating biomarkers.¹¹⁴ They can be used for diagnosing diseases and predicting their outcomes. Moreover, they also hold potential as targets for cancer treatment.

Figure 4.

Mechanisms of snoRNAs in Colorectal Cancer. (Created in BioRender)

(A) Affects ribosome function.

(B) Regulation of p53 protein.

(C) Regulation of signaling pathways.

(D) Other potential modes of regulation.

Table 1.

Functions of small nucleolar RNAs in colorectal cancer

snoRNA	Type	Expression	Mechanism	Potential functions	Reference
SNORA5C	H/ACA	Up-regulation	Unknown	Promote proliferation and colony formation	Shen et al.98
SNORA15	H/ACA	Up-regulation	Unknown	Promotes carcinogenesis and participates in inflammation	Yang et al.99
SNORA21	H/ACA	Up-regulation	Hippo signaling pathway, and Wnt signaling	Promote cell proliferation, and enhance tumor invasiveness	Yoshida et al.100
SNORA24	H/ACA	Up-regulation	p53	Promote G1/S phase transition and cell proliferation.	Shen et al.101
SNORA28	H/ACA	Up-regulation	JAK-STAT pathway	Promotes Cell Proliferation and Radioresistance	Liu et al.102
SNORA33	H/ACA	Up-regulation	Unknown	Promotes vascular invasion	Yuan et al.103
SNORA41	H/ACA	Up-regulation	Unknown	Promotes carcinogenesis and participates in inflammation	Yang et al.99
SNORA56	H/ACA	Up-regulation	Pseudouridine modification of rRNA inhibits ferroptosis	Promote cell proliferation	Xu et al.92
SNORA71A	H/ACA	Up-regulation	LBP, NF-κB, Toll-like receptor	Promote cell proliferation, migration and invasion	Zhang et al.104
SNORD1C	C/D	Up-regulation	Wnt/β-catenin pathway	Promote cell proliferation, migration, invasion, and enhanced cancer cell stemness	Liu et al.105,106
SNORD12B	C/D	Up-regulation	Unknown	Promote tumorigenesis, proliferation, and metastasis	Xu et al.107
SNORD12C	C/D	Up-regulation	ZFAS1-NOP58-SNORD12C/78-EIF4A3/LAMC2	Promote cell proliferation	Wu et al.108
SNORD15B	C/D	Up-regulation	Unknown	Promote proliferation and colony formation	Shen et al.98
SNORD19	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD33	C/D	Down-regulation	Unknown	Suppress cell anchorage	Yang et al.99
SNORD44	C/D	Down-regulation	Caspase dependent pathway, PI3K/Akt	Inhibit tumor growth, and induce apoptosis	Yuan et al.110
SNORD50A	C/D	Down-regulation	Methylation of 28S rRNA gene, k-Ras	Promote tumor growth	Su et al.70
SNORD50B	C/D	Down-regulation	Methylation of 28S rRNA gene, k-Ras	Promote tumor growth	Su et al.70
SNORD59A	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD63	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD63B	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD78	C/D	Up-regulation	ZFAS1-NOP58-SNORD12C/78-EIF4A3/LAMC2	Promote cell proliferation	Wu et al., Okugawa et al.108,111
SNORD99	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD100	C/D	Down-regulation	Unknown	Influences the inflammatory microenvironment	Cai et al.109
SNORD126	C/D	Up-regulation	PI3K/Akt	Promote cell growth	Fang et al.112
SCARNA12	scaRNA	Up-regulation	PI3K/Akt	Promote proliferation and tumorigenicity	Zhang et al.113

Affects ribosome function

As the principal site of polypeptide biogenesis within the cellular environment, the ribosome is composed of approximately 80 distinct ribosomal proteins in conjunction with four rRNA species (5S, 5.8S, 18S, and 28S). Numerous data have demonstrated that changes in ribosome activity are usually associated with cellular carcinogenesis.¹¹⁵ Recent studies have found that aberrant snoRNA-mediated rRNA modifications are closely associated with CRC progression, and the mechanism of action may involve the disruption of the ribosome biosynthesis regulatory network, thereby promoting the proliferation and invasion of tumor cells.

LncRNA ZFAS1 recruits NOP58 to assemble SNORD12C/78 snoRNP, catalyzing 28S Gm3878/4593 Nm and boosting the translation of oncogenes (EIF4A3, LAMC2) to drive CRC proliferation and invasion.¹⁰⁸ However, the study is still confined to cell lines and xenograft models; direct validation of pathway activity in clinical specimens and its association with patient prognosis are lacking, leaving its translational potential uncertain.

SNORD11B has been reported to possess dual functions: mediating Nm at the G509 site of 18S rRNA to promote ribosomal maturation, and facilitating degradation by mediating Nm modification at the G225 site of pri-let-7a, thereby downregulating the tumor suppressor miRNA let-7a-5p and certain oncogenes.¹¹⁶ However, the core premise of this regulatory model—the Nm modification on pri-let-7a itself—remains urgently in need of independent and rigorous validation across diverse experimental systems.

Xu et al. discovered that SNORA56 catalyzes Ψ at U1664 in the 28S rRNA, thereby upregulating GCLC protein expression, enhancing glutathione synthesis, and inhibiting lipid peroxidation. This process inhibits ferroptosis and enhances the growth and malignant potential of CRC.⁹² Its capacity to induce ferroptosis, however, was mainly observed in specific cell types, such as HT29 and HCT8, suggesting that the mechanism likely relies on distinct metabolic features of different cells. Therefore, the extent to which this finding applies requires further verification.

Current findings point to a critical role for snoRNA-guided rRNA modifications in CRC. Subsequent research should employ single-cell sequencing and organoid cultures to dissect tumor heterogeneity and assess their potential for clinical diagnosis and therapy.

Regulation of p53 protein

Functioning as a potent tumor suppressor, post-translational modifications activate p53 and orchestrate key anti-tumor defenses, including cell-cycle arrest, apoptosis, and senescence.¹¹⁷ Beyond these transcriptional roles, p53 also places a direct check on ribosome biogenesis by regulating FBL to inhibit rRNA methylation.¹¹⁸ Notably, emerging evidence has identified specific snoRNAs as critical modulators within the p53 regulatory network.^70,101,119

SNORA24 demonstrates divergent, cancer-specific roles. While its expression is reduced in HCC, implying a tumor-suppressive function, it is notably elevated in CRC and other malignancies.¹²⁰ In these contexts, higher SNORA24 levels are strongly linked to more aggressive tumor characteristics.¹²⁰

SNORA24 drives CRC progression by degrading the p53 tumor suppressor. This action, demonstrated by Shen et al., is achieved by directing p53 for proteasomal destruction.¹⁰¹ The subsequent loss of p21 and other target genes releases cells from cycle arrest, expedites the G1/S transition, and ultimately underpins the observed suppression of apoptosis and enhancement of proliferative and clonogenic potential.¹⁰¹

Nevertheless, this study still faces particular unaddressed difficulties. For example, it has not clarified the precise molecular mechanism by which SNORA24 regulates this process, while the number of studies worldwide on the interaction between snoRNAs and the p53 protein to influence CRC development is extremely low.

Regulation of signaling pathways

In CRC, classical signaling pathways such as the Wnt/β-catenin, TGF-β, PI3K/Akt, and MAPK/ERK pathways are either aberrantly activated or inhibited. The malfunction of these pathways can remarkably control the malignant characteristics of tumor cells. Recent research advancements have revealed that diverse snoRNAs exert critical regulatory effects on colorectal tumor development by directly or indirectly modulating signaling networks.

Multiple studies suggest that snoRNAs exert widespread regulatory effects on the PI3K/AKT pathway. For example, SNORD126 upregulates FGFR2, which triggers the PI3K-AKT signaling cascade. This activation is marked by the increased phosphorylation of AKT (at Thr308 and Ser473) and its downstream effectors GSK-3β and p70S6K, ultimately accelerating the proliferation of CRC cells.¹¹² SCARNA12 appears to function similarly, as its growth-promoting effects can be blocked by AKT inhibitors—a conclusion supported by RNA-seq and KEGG pathway analysis.¹¹³

Regarding the Wnt/β-catenin pathway, studies have shown that SNORD1C promotes Wnt signaling activation.¹⁰⁵ This is achieved by increasing β-catenin protein expression and facilitating its complex formation with the transcription factor TCF7.¹⁰⁵ Together, they act synergistically to drive the transcription of stemness-related genes such as MYC and SOX2.¹⁰⁵ As a result, tumor cells exhibit enhanced stemness, increased proliferative capacity, and greater resistance to treatment.^105,106,119 Although this mechanistic model presents certain novel aspects, it still lacks evidence of actual Wnt pathway activation in clinical samples. Further validation is needed to assess its broader applicability and potential for clinical translation.

In contrast, certain snoRNAs, such as SNORA28, have been studied in greater depth. Research has revealed that they can recruit bromodomain protein 4 (BRD4) to the leukemia inhibitory factor receptor (LIFR) promoter through epigenetic regulation, thereby activating the JAK-STAT signaling pathway and enhancing radiation resistance in CRC cells.¹⁰²

SNORD44—another snoRNA—exhibited an anticancer effect. It synergizes with GAS5 to activate Caspase3/8/9, which specifically cleaves PARP, thereby inactivating its DDR activity and inhibiting tumor cell growth.¹¹⁰

Additionally, SNORD50 A/B deletion is prevalent in many cancers, especially CRC. The loss of SNORD50 A/B has been reported to be associated with K-Ras signaling activation; however, the lack of direct experimental validation means that the causal relationship with KRAS mutations remains unclear.⁷⁰ Similarly, associations of SNORA21 and SNORA71A with Hippo and NF-κB pathways—primarily derived from bioinformatics analyses—await functional validation to define their precise molecular roles and downstream targets in CRC.¹²¹

Other potential modes of regulation

In addition to the regulatory mechanisms stated above, several additional snoRNAs have been linked to specific characteristics of CRC. Downregulation of SNORD33 expression was reported to limit CRC cells’ anchoring capacity.⁹⁹ SNORA5C and SNORD15B were found to be tightly linked with CRC cell proliferation.⁹⁸

Several expression profiling studies have suggested potential roles for specific snoRNAs in CRC pathogenesis. Yang et al. reported significant upregulation of SNORA15 and SNORA41, accompanied by the downregulation of SNORD33, in CRC tissues, with expression levels following a progressive trend across the “healthy tissue–ulcerative colitis–CRC” sequence.⁹⁹ This pattern was further corroborated by Yuan et al., who independently identified elevated expression of SNORA33 in CRC, reinforcing the notion that certain snoRNAs are consistently dysregulated during colorectal carcinogenesis.¹⁰³ Nevertheless, these findings remain largely correlative, constrained by limited cohort sizes and a lack of mechanistic validation, leaving the functional contributions of these snoRNAs to CRC progression incompletely resolved.

Notably, the functional scope of snoRNAs may not be limited to rRNA modifications. Cai et al. proposed that a specific set of snoRNAs (termed “TIIsno”) can influence the tumor immune microenvironment, as their expression profiles negatively correlate with the levels of immune cell infiltration and checkpoint expression.¹⁰⁹ However, this model is mainly based on computational inference, and the cellular origin of these snoRNAs and their specific mechanisms of action on immune function have not been directly experimentally verified.

In CRC research, the expression levels of specific snoRNAs, such as SNORD12B, have shown significant heterogeneous characteristics.¹¹⁶ The expression patterns of these snoRNAs are significantly correlated with complex tumor phenotypes, including a hypoxic microenvironment and metabolic reprogramming.¹⁰⁷ However, additional functional experiments are still needed to verify these statistical associations and clarify their causal mechanisms.

Overall, some findings are based on correlation analyses using limited samples, and their generalizability requires validation in large-scale cohort studies. Future research should employ more refined experimental models to translate statistical associations into mechanistic understanding and assess their potential for clinical translation.

Clinical implications and therapeutic potential of small nucleolar RNAs in colorectal cancer

Small nucleolar RNAs as liquid biopsy biomarkers

Recent studies have shown that snoRNA can be encapsulated in extracellular vesicles such as exosomes and enter the bloodstream through the circulatory system.^{122,123,124,125} The vesicle structure effectively protects snoRNA from nuclease degradation, thereby significantly increasing its stability in plasma. This property, together with the dual regulatory roles of snoRNA in CRC, provides a solid biological basis for its use as a liquid biopsy marker.¹¹⁴

Available clinical data provide strong support for this theory. As summarized in Table 2, specific serum snoRNAs have demonstrated excellent diagnostic capabilities in some cancer types, with performance comparable to or even superior to that of the traditional marker CEA. In the field of CRC, SNORD11B, with an AUC of 0.886, is a prominent example.¹¹⁶ It is especially noteworthy that the combination of snoRNA and CEA can produce a significant synergistic effect. In early CRC screening, the combined diagnostic model constructed from SNORA56 and CEA increased the AUC to about 0.92, achieving a qualitative leap in diagnostic efficacy.⁹² This breakthrough indicates that the clinical value of snoRNA lies not only in its use as an independent diagnostic index, but, more importantly, in its ability to complement the existing diagnostic system and improve diagnostic efficacy together.

Table 2.

Diagnostic performance of snoRNAs in serum in some tumors

Cancer	snoRNA	AUC	SPE	SEN	AUC of CEA	AUC of CEA+snoRNA	Reference
CRC	SNORD1C	0.7480	57.45%	79.80%	0.7150	0.8380	Liu et al.106
CRC	SNORA56	0.7572	/	/	0.8901	0.9169	Xu et al.92
CRC	SNORD11B	0.8862	/	/	0.8257	0.9582	Bian et al.116
CRC	SNORA33	0.6250	79.00%	45.00%	0.5450	0.9070	Yuan et al.103
ESCA	SNORA58, SNORA68, SNORD93	0.8570	72.70%	86.20%	0.7150	0.8780	Zhang et al.126
ccRCC	SNORD96A	0.8909	80.00%	27.30%	/	/	Shang et al.127
BC	SNORD16	0.7334	61.09%	76.75%	/	/	Li et al.6
BC	SNORA73B	0.7165	67.80%	68.38%	/	/	Li et al.6
BC	SNORD49B	0.6728	70.22%	57.39%	/	/	Li et al.6
BC	SCARNA4	0.6880	74.22%	58.37%	/	/	Li et al.6
NSCLC	SNORD66	0.8139	77.27%	75.68%	/	/	Liao et al.128
NSCLC	SNORD76	0.8064	90.91%	70.27%	/	/	Liao et al.128
NSCLC	SNORD33	0.8233	86.36%	72.97%	/	/	Liao et al.128
NSCLC	SNORA83	0.7016	52.40%	84.7%	0.6585	0.7453	Wang et al.129

SEN, sensitivity; SPE, specificity; CRC, colorectal cancer; ESCA, esophageal carcinoma; ccRCC, clear cell renal cell carcinoma; BC, breast carcinoma; NSCLC, non-small cell lung cancer.

The diagnostic potential of snoRNA in different samples is pushing it to gradually move toward the clinical practice of liquid biopsy (Table 3). Although tissue testing (such as SNORA21, AUC = 0.9300) is still the gold standard for diagnosis and can trace the source of the tumor of circulating RNA, its trauma limits clinical application.^100,130 The snoRNA in the blood sample (such as SNORD1C, AUC = 0.7480) shows unique advantages: non-invasive sampling makes it suitable for early screening, follow-up of high-risk people, and dynamic monitoring of treatment effect, and the characteristics of repeated sampling can also effectively avoid the deviation caused by tumor heterogeneity.^106,131 It is worth noting that snoRNAs such as SNORA56 and SNORD11B maintain stable diagnostic performance in tissues and blood samples.^92,116 This discovery closely links circulating snoRNA with primary tumors and provides a solid biological basis for their application in blood testing. At present, RNA-FISH, RT-PCR, Northern blotting, RNA sequencing, RNA microarray, and other mature technology platforms provide strong support for the detection of such molecules.^{132,133,134,135,136,137}

Table 3.

Diagnostic properties of snoRNAs in colorectal cancer tissues and plasma

snoRNA	Sample	AUC	Reference
SNORD1C	Blood	0.7480	Liu et al.106
SNORA56	Blood	0.7572	Xu et al.92
SNORD11B	Blood	0.8862	Bian et al.116
SNORA33	Blood	0.6250	Yuan et al.103
SNORA56	Tissue	0.6759	Xu et al.92
SNORD11B	Tissue	0.6784	Bian et al.116
SNORA28	Tissue	0.6946	Liu et al.102
SCARNA12	Tissue	0.8597	Zhang et al.113
SNORA71A	Tissue	0.7847	Zhang et al.104
SNORA21	Tissue	0.9300	Yoshida et al.100
SNORA24	Tissue	0.6695	Shen et al.101

However, snoRNA still faces challenges to achieve clinical standardization. The first problem is the imbalance of diagnostic performance. Although the sensitivity of SNORD1C is 79.80%, the specificity is only 57.45%; the specificity of SNORA33 is 79.00%, but the sensitivity is as low as 45.00%.^103,106 This characteristic makes it difficult for a single marker to undertake diagnostic tasks independently, and it needs to be considered in conjunction with CT, MRI, and other imaging examinations. In addition, the whole process of liquid biopsy (from sample processing to RNA detection) has not yet established a unified standard, which not only puts forward higher requirements for laboratory operation specifications, but also makes it difficult to repeat the results between laboratories, and the risk of operational pollution increases the possibility of false positives.

Small nucleolar RNAs as prognostic biomarkers

The potential of snoRNAs in the prognostic assessment of CRC is receiving increasing attention (Table 4). Based on their associated biological behaviors, they can be classified into several functional categories. Among them, high expression of SNORA28, SNORA71A, and SNORA21 is closely associated with invasive features such as lymphatic metastasis and venous invasion, suggesting that they may play an important role in molecular pathways regulating tumor cell migration and infiltration.^100,102,104 However, the research data for SNORA28 and SNORA71A were mainly obtained from the TCGA public database and need to be validated in independent clinical samples.^102,104

Table 4.

Prognosis performance of snoRNAs in colorectal cancer

snoRNA	Characteristitis	Prognosis in patients with high expression	Reference
SNORA28	Lymphatic invasion, Venous invasion, History of colon polyps	Poor	Liu et al.102
SCARNA12	History of colon polyps	Poor	Zhang et al.113
SNORD1C	Tumor infiltration, CEA (ng/mL)	Poor	Liu et al.106
SNORA71A	TNM stage, Lymph node metastasis	Poor	Zhang et al.104
SNORA21	Age, Tumor invasion, Distant metastasis, TNM stage	Poor	Yoshida et al.100
SNORD126	/	Poor	Fang et al.112
SNORD11B	/	Poor	Bian et al.116
SNORA56	/	Poor	Xu et al.92
SNORA24	Age, History of colon polyps	Poor	Shen et al.101

Another class of snoRNAs, such as SNORA24 and SCARNA12, has been associated with precancerous lesions, and their high expression is often observed in patients with a history of colon polyps, which may represent early molecular events during malignant transformation.^101,113 However, especially for SCARNA12, the available clinical data remain limited, and the sample size needs to be expanded to more deeply explore its specific function in colorectal carcinogenesis.¹¹³

In addition, the expression levels of some snoRNAs, such as SNORD1C and SNORA21, are correlated with classical clinical indicators, such as CEA and TNM stage, suggesting that more accurate risk prediction models could be constructed if combined with traditional prognostic parameters.^100,106 However, this type of study still has limitations, such as the small sample size of SNORD1C and the uneven distribution of patient stages across cohorts in the analysis of SNORA21, which need to be optimized in subsequent studies.^100,106

Finally, snoRNAs such as SNORD126, SNORD11B, and SNORA56 are associated with poor prognosis, but their association with specific clinicopathologic features has not yet been clarified, and the related mechanisms remain to be further explored.^92,112,116

Therapeutic opportunities and challenges

In the therapeutic field, targeting abnormally expressed snoRNAs has emerged as a highly promising strategy, with CRISPR/Cas9 and antisense oligonucleotides (ASOs) serving as core technologies. In CRC, intervention strategies are shifting from single functional modulation to multi-pathway synergistic therapy. Direct functional interventions have achieved remarkable results: knockdown of oncogenic SNORA21 by CRISPR-Cas9 technology can effectively inhibit tumor growth¹⁰⁰; while delivery of oncogenic SNORD44 using lysogenic adenovirus vectors successfully induced apoptosis in tumor cells.¹¹⁰ More groundbreakingly, the interaction between the snoRNA regulatory network and emerging cell death pathways (e.g., iron death) opens up new directions for the development of synergistic therapies. Experimental studies have demonstrated that SNORA56 inhibits the process of iron death through the “28S rRNA-GCLC” molecular axis, and when combined with the ferroptosis inducer RSL3, it produces a potent synergistic anti-tumor effect, increasing the tumor suppression rate to 78.6%.⁹² Antisense oligonucleotides (ASOs), as another core technology, show significant value in snoRNA-targeted therapy. Studies have shown that ASOs can effectively target SNORD14E, providing a new potential therapeutic pathway for endometrial cancer.¹³⁸

Although the aforementioned preclinical studies have shown promising results, these findings remain primarily limited to experimental models. In the more complex human tumor microenvironment, their efficacy and safety remain to be verified. For example, in ASO therapy, immunogenic vectors, chemical modifications, and drug accumulation may trigger systemic toxicity and lead to liver and kidney injury.¹³⁹ In addition, the high degree of sequence homology within snoRNA gene clusters (e.g., SNORD115/116) poses a challenge for the precise design of ASO/siRNAs, which not only increases the risk of off-targeting but may also disrupt the entire genetic regulatory network.¹⁴⁰

Future advances require tissue-specific delivery systems, optimized chemical modifications, and rigorous target validation to ensure both efficacy and safety.

Conclusion and prospects

SnoRNA has undergone functional evolution in CRC, changing from a classic butler molecule to a multifunctional regulatory factor with important research and clinical potential. Their typical role involves guiding the chemical modification of the ribosomal RNA in the nucleus, such as methylation and pseudouracinization - a process that is crucial to the assembly and function of the ribosome. The disruption of these activities can impair ribosomal integrity and protein synthesis, resulting in loss of cell homeostasis.¹⁴¹

Outside the nucleus, snoRNA shows complex regulatory behavior in the cytoplasm. Through interaction with mRNA, proteins, and non-coding RNA, including tRNA,^81,142 they affect translation efficiency, signal transduction, and stress adaptation, thus playing an important role in tumor development and metabolic diseases. In CRC, studies have shown that several cytoplasmic snoRNAs can regulate the expression of key oncogenes and tumor suppressor genes, which highlights their therapeutic relevance.

Despite these insights, the fundamental problems still exist. The mechanism regulating the selective nuclear export of snoRNAs remains unclear, and it is not clear whether their presence in the cytoplasm results from active transport or passive leakage - this distinction is of important functional significance. In addition, the potential competition between classical snoRNA functions and non-classical snoRNA functions has not been systematically studied.

Current technical limitations limit target identification and functional verification, which makes many snoRNA regulatory networks under-understood. In addition, their high tissue specificity and space-time expression patterns make it complicated to draw a comprehensive pathological expression spectrum. The most important thing is that the current research is mainly at the level of molecular expression correlation (as shown in Table 1) and lacks systematic clarification of the snoRNA dysfunction drivers in CRC, its downstream functional effects, and the causal relationships between these and disease phenotypes.

The latest progress of high-resolution sequencing technology has significantly increased our understanding of the rRNA modified landscape in cancer.^143,144,145 At present, technologies such as Nanopore direct RNA sequencing, RiboMeth-seq, and HydraPsiSeq can quantify a variety of rRNA modifications (including Nm and Ψ) in clinical tumor samples.^143,144,145 In the future, these advanced technologies are expected to systematically map snoRNA expression and modification across a large-scale CRC cohort. This method will enable the discovery of new pathogenic mechanisms and the definition of biomarker combinations with excellent diagnostic performance to be realised simultaneously, with all information derived from a single data source.

New biological dimensions are also emerging. More and more evidence indicates that SNORA51 has been detected in fecal samples, suggesting a potential “snoRNA-intestinal microbiome” axis in CRC.¹⁴⁶ Key issues remain, including whether specific intestinal microorganisms or metabolites (such as short-chain fatty acids) regulate the expression of colon snoRNA, and whether tumor-derived exocrine snoRNA is ingested by symbiotic bacteria or immune cells, thereby reshaping the tumor microenvironment and microbial ecology. These unresolved problems provide key guidelines for future research direction. In addition, integrating snoRNA biomarkers verified by liquid biopsy with patients’ genome and transcriptome data is expected to reveal their deep relationships with specific CRC subtypes and provide important evidence for the development of precise treatment strategies based on snoRNA.

It is known that some tumors can use exocrine bodies for intercellular communication, shape the pre-metastasis microenvironment, and escape immune attack.^{147,148,149,150} In addition, research shows that snoRNA may be packaged in the exocrine body and released into the bloodstream through exocrine vesicles (such as exocrine bodies).^{122,123,124,125} The potential connection between exocrine, snoRNA, and the tumor microenvironment suggests the possibility of developing a new treatment method: the targeted secretion of cancer-promoting snoRNA. By developing strategies to inhibit the loading and release of these exocrine bodies containing snoRNA, we can inhibit the growth of CRC tumors through mechanisms different from traditional therapies.

Acknowledgments

This work was supported by grants from the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX25_3804), the National Natural Science Foundation of China (81972015), the Social Development Project from the Jiangsu Provincial Department of Science and Technology (BE2020770), the Jiangsu Provincial Research Hospital (YJXYY202204-YSB72), and the 6th phase of Nantong Jianghai Talent Program.

Finally, We thank Figdraw (http://www.figdraw.com) and biorender (https://www.biorender.com/) for providing the platform used to create the figures in our article.

Author contributions

Z.Y. was involved in the study’s conception and design. Y.C. and S.L. were responsible for material preparation, data collection, and analysis. Z.Y. wrote the initial draft of the article. C.T. contributed to the proofreading of the article. X.H. and J.Y. participated in the article revision and redrawing of illustrations. X.W. and S.W. provided resources and guidance throughout the study. All authors reviewed and commented on previous versions of the article and approved the final version.

Declaration of interests

The authors declare no competing interests.

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

During the preparation of this work, the authors used Grammarly and DeepL Write in order to assist with language polishing, grammar checking, and sentence refinement. After using these tools, the authors thoroughly reviewed and edited the content as needed and take full responsibility for the content of the publication.