Small Nucleolar RNA Snord17 Promotes Self‐Renewal of Intestinal Stem Cells through Yy2 mRNA Export and Tead4 Activation

ABSTRACT

The intestinal epithelium possesses a profound capacity for regeneration, which is fueled by the proliferation and differentiation of leucine‐rich repeat‐containing G‐protein‐coupled receptor 5 (Lgr5)‐expressing intestinal stem cells (ISCs) located at the base of the crypts. However, how small nucleolar RNAs (snoRNAs) regulate the self‐renewal of ISCs remains elusive. Here, we identified a small nucleolar RNA Snord17 that is highly expressed in ISCs. Snord17 knockout abrogates stemness of ISCs and impairs epithelial regeneration. Mechanistically, Snord17 interacts with THO complex 3 (Thoc3) to facilitate nuclear export of Yin yang 2 transcription factor (Yy2) mRNA for its subsequent translation. Yy2 protein enriches on the promoter of TEA domain family member 4 (Tead4) to activate its transcription, leading to activation of Hippo signaling for self‐renewal maintenance of ISCs. Of note, Tead4 deficiency impairs self‐renewal of ISCs and intestinal regeneration. Our findings reveal that the Snord17‐Thoc3‐Yy2‐Tead4 axis is required for self‐renewal maintenance of ISCs and gut regeneration.

Snord17, through interaction with Thoc3, promotes nuclear export and translation of Yy2 mRNA in Snord17 ^+/+ ISCs. The Yy2 protein subsequently binds the Tead4 promoter to promote its transcription, activating Hippo signaling, which is essential for ISC maintenance. In Snord17 ^−/− ISCs, impaired Yy2 mRNA export prevents translation, leading to suppressed Tead4 expression and disrupted Hippo signaling.

1. Introduction

The intestinal epithelium plays a pivotal role in nutrient absorption, barrier defense, and immune regulation, which relies on rapid regeneration driven by intestinal stem cells (ISCs)​​ positioned in the crypt niche [1, 2]. The ISCs harbor remarkable self‐renewal and differentiation ability, and perpetually generate transit‐amplifying (TA) progenitors that ultimately differentiate into five major cell lineages, including tuft cells, enterocytes, goblet cells, Paneth cells, and enteroendocrine cells [3, 4]. ISCs are regulated by several signaling pathways in the niche, including Wnt/β‐catenin, Notch, BMP, Hippo, and GABA signaling [5, 6, 7]. These signaling pathways engage in ​​synergistic crosstalk​​ to orchestrate the self‐renewal of ISCs, while dysregulation of these signaling pathways disrupts intestinal epithelial renewal and impairs injury repair, which contributes to the pathogenesis of inflammatory bowel diseases (IBD) and colorectal cancer (CRC) [8, 9]. However, how the self‐renewal maintenance of ISCs is regulated remains incompletely understood.

SnoRNAs are a conserved class of noncoding RNAs, typically ranging from 60 to 300 nt in length and predominantly found in the nucleolus [10]. Based on sequence and secondary structure, snoRNAs are generally classified into two major types: ​​H/ACA box snoRNAs​​ and ​​C/D box snoRNAs​​, which guide ​​pseudouridylation​​ and ​​2'‐O‐methylation​​, respectively [11]. While snoRNAs are classically characterized for their roles in post‐transcriptional modification and maturation of rRNAs by forming small nucleolar ribonucleoproteins (snoRNPs), emerging studies demonstrates their non‐canonical function in diverse biological processes, including epigenetic regulation, protein secretion and disease pathogenesis [12]. For instance, SNORA73 binds with mRNA and 7SL RNA to enhance targeted mRNA engagement with signal recognition particle (SRP), thereby facilitating secretion of secretory and membrane proteins [13]. SNORD88B promotes self‐renewal of liver cancer stem cells by anchoring Werner Syndrome Protein (WRN) in the nucleolus to suppress serine/threonine kinase 4 (STK4) transcription [14]. ​​However, it is still unclear​​ how snoRNA govern the underlying mechanisms responsible for ISCs function.

mRNA nuclear export for translation represents a critical biological process [15, 16], which is regulated by a transcription‐export (TREX) complex. It's reported that TREX facilitates aggregation of mature mRNAs within nuclear speckles, enabling rapid cytoplasmic translocation upon transcriptional reactivation to sustain cellular viability [17]. Perturbation of TREX components disrupts nuclear export of mRNAs encoding neurotransmitters in neurons and stemness‐associated genes in triple‐negative breast cancer, impairing neuronal function and tumorigenic properties [18, 19]. ​​ However, it remains unknown how TREX components‐mediated mRNA export regulates the self‐renewal capability of ISCs.

Tead4, a key mediator of the Hippo signaling pathway, forms a complex with Yes‐associated protein 1/Transcriptional coactivator with PDZ‐binding motif (Yap/Taz) to activate downstream genes to support cellular proliferation, survival, and stemness [20, 21]. Tead4 partners with Vgll1 modulate chromatin accessibility at target gene loci through histone acetylation to promote cell proliferation, which is critical for proper embryonic development [22]. The activation of Tead4 via glucocorticoid receptor (GR) signaling promotes the maintenance of breast cancer stem cells (BCSCs), closely associating with tumor chemoresistance [23]. However, it is unclear how Tead4 is involved in the ISC regulation.

In order to explore the role of snoRNAs in regulation of ISC stemness, we identified a snoRNA Snord17 that was highly expressed in ISCs. Snord17 interacted with TREX component Thoc3 to facilitate the nuclear export of Yy2 mRNA for translation. Yy2 initiated the transcription of Tead4, thereby sustaining ISC self‐renewal and intestinal homeostasis by activation of the Hippo signaling pathway. Of note, this SNORD17‐THOC3‐YY2‐TEAD4 regulatory axis was aberrantly regulated in both IBD and CRC patients, implicating its potential contribution to gut disease pathogenesis.

2. Results

2.1. Snord17 is Highly Expressed in ISCs

To investigate the role of snoRNAs in the self‐renewal of ISCs, we isolated Lgr5⁺ ISCs and Lgr5⁻ intestinal epithelial cells (IECs) from intestinal crypts of Lgr5 ^{GFP‐CreERT2} mice for subsequent snoRNA transcriptome sequencing. We identified 10 upregulated snoRNAs in ISCs compared to IECs, and Snord17 exhibited the most differential expression (Figure 1A). Actually, Snord17 showed the highest expression in ISCs among the above top 10 snoRNAs (Figure 1B). Subsequently, we employed short hairpin RNA (shRNA)‐mediated knockdown of Snord17 in intestinal organoids, and Snord17 depletion obviously suppressed organoid formation (Figure 1C; Figure S1A). Snord17 was generated from the intron located between exon2 and exon3 of sorting nexin 5 (Snx5), with a length of 238 nt, and conserved across various species (Figure S1B–D). As predicted by secondary and tertiary structural analysis, Snord17 exhibited a canonical C/D box structure in its 5' and 3' ends, consistent with defined characteristics of snoRNAs (Figure S1E,F). RNA fluorescence in situ hybridization (RNA‐FISH) analysis revealed pronounced enrichment of Snord17 in the intestine and liver (Figure 1D), which was further validated by Northern blotting (Figure 1E). In addition, Snord17 was highly expressed in intestinal crypt niches, showing elevated expression in Lgr5⁺ ISCs compared to Lgr5⁻ IECs (Figure 1F; Figure S1G,H). Moreover, Snord17 exhibited specific enrichment at bud‐forming tips of intestinal organoids, recapitulating its localization in crypt regions of small intestines (Figure 1G). Similar to the small intestine, high expression of Snord17 was also observed in Lgr5⁺ stem cells residing in the colon (Figure S1I).​ Further subcellular fractionation and RNA‐FISH in ISCs showed main localization of Snord17 in nucleoli and nucleoplasm (Figure 1H,I). Taken together, Snord17 is highly expressed in the nucleoli and nucleoplasm of Lgr5⁺ ISCs.

FIGURE 1.

Snord17 is highly expressed in ISCs. (A) Lgr5⁺ and Lgr5⁻ cells were sorted from intestinal crypts of Lgr5 ^{GFP‐CreERT2} mouse using FACS, followed by SnoRNA transcriptome analysis. The heatmap displays snoRNAs with high expression in Lgr5⁺ cells, and the top 10 snoRNAs are denoted. (B) Relative expression of the top10 snoRNAs in Lgr5⁺ ISCs via qRT‐PCR analysis. n = 4 mice for each group. (C) Snord17 was knocked down in isolated Lgr5⁺ ISCs using shRNA, followed by organoid formation. Representative images of shCtrl and shSnord17 organoids are shown in the left panel, and organoid formation ratios are shown in the right panel. n = 6 independent experiments for each group. Scale bars, 50 µm. (D) One‐week‐old mice were euthanized for longitudinal sections. A global view of the section is shown in the left panel, and the indicated tissues are shown in the right panel. 1, brain; 2, heart; 3, lung; 4, liver; 5, intestine; 6, kidney. Scale bars, 400 µm. (E) Northern blotting analysis of Snord17 in different tissues. 18S rRNA served as a loading control. (F) RNA‐FISH of Snord17 in jejunum sections. Scale bars, 100 µm. (G) RNA‐FISH of Snord17 in organoids. Scale bars, 50 µm. (H) Total RNA from Lgr5⁺ ISCs was extracted from the cytoplasm, nucleoplasm, and nucleolus, followed by qRT‐PCR (left panel) and Western blotting (right panel). U1 RNA served as a positive control for nuclear location. β‐actin, H3, and Nucleolin are protein markers for cytoplasm, nucleoplasm, and nucleolus, respectively. n = 4 independent experiments. (I) Snord17 was visualized by RNA‐FISH, followed by immunofluorescence staining in ISCs. Scale bars, 5 µm. Data are shown as the means ± SD. Statistical analysis was performed using unpaired two‐tailed Student's t‐tests. ^*** p < 0.001.

2.2. Snord17 Knockout Impairs Self‐Renewal of ISCs and Intestinal Regeneration

To further determine the physiological role of Snord17 in ISCs, we generated Snord17 knockout mice using CRISPR/Cas9 technology (Figure S2A). Deletion of Snord17 was validated by genotyping, qRT‐PCR, and Northern blotting (Figure S2B–D). ​​Although Snord17 resided within the intronic region of Snx5 gene locus on mouse chromosome 2, CRISPR/Cas9‐mediated deletion of Snord17 had ​​no impact on Snx5 expression (Figure S2E,F).

We found that Snord17 ^−/− mice showed shorter length of small intestines and colons compared to littermate Snord17^+/+ control mice (Figure 2A; Figure S2G). The impairment of stem cell stemness has been reported to cause shortened crypt and villus, which in turn reduces the digestive and absorptive capacity of the small intestine [24]. Small intestines from Snord17 ^−/− mice displayed shortened villi and crypts, with reduced crypt number compared to control mice (Figure 2B; Figure S2H,I). ​​In addition, Snord17 deficiency also caused shortened colonic crypts (Figure S2J). Meanwhile, elevated levels of glucose and non‐esterified fatty acid were detected in the feces of Snord17 ^−/− mice, indicating impaired nutrient absorption (Figure S2K). We generated Lgr5 ^{GFP‐CreERT2};Snord17 ^−/− mice to enable GFP‐based labeling of ISCs in Snord17^−/− mice. Consistently, Snord17 ^−/− mice exhibited decreased Lgr5⁺ ISCs in small intestines (Figure 2C). Reduced ISCs were further confirmed by analyzing other markers, such as olfactomedin 4 (Olfm4), Bmi1 polycomb ring finger oncogene (Bmi1), SPARC related modular calcium binding 2 (Smoc2), and 3‐hydroxy‐3‐methylglutaryl‐Coenzyme A synthase 2 (Hmgcs2) (Figure 2D; Figure S2L). In addition, Ki67 staining revealed remarkable decreased proliferating cells in the small intestines and colons of Snord17 ^−/− mice (Figure 2E; Figure S2M). Subsequently, intestinal crypts were isolated from Snord17^+/+  and Snord17 ^−/− mice, followed by organoid‐forming assay and Ki67 staining. In parallel, organoids from Snord17 ^−/− mice displayed decreased proliferating cells (Figure 2F), while cleaved caspase3 staining showed no difference in cell apoptosis between Snord17^+/+ and Snord17 ^−/− organoids (Figure 2G). Given an established role of Lgr5⁺ ISCs in injured epithelium regeneration [25], we evaluated intestinal damage repair and crypt regeneration in Snord17‐deficient mice. Following 8 Gy irradiation, intestinal tissues from Snord17 ^−/− mice displayed impaired epithelia repair, indicating impaired regenerative capacity of ISCs upon Snord17 deletion (Figure 2H). Meanwhile, organoid formation of Snord17 ^−/− ISCs was abrogated, whereas reintroducing Snord17 in Snord17 ^−/− ISCs successfully rescued organoid formation ability (Figure 2I; Figure S2N). Altogether, these data indicate that Snord17 participates in the enhancement of stemness maintenance of ISCs and intestinal regeneration.

FIGURE 2.

Snord17 knockout impairs self‐renewal of ISCs and intestinal regeneration. (A) Images of small intestines and colons from Snord17^+/+ and Snord17 ^−/− mice are shown. Length of small intestines are calculated. n = 4 mice for each group. (B) Representative H&E staining images of jejunum sections from Snord17^+/+ and Snord17 ^−/− mice. Scale bars, 100 µm. Length of villus, crypt, and crypt number per field are shown in the right panel. n = 32 villi for each group. n = 30 crypts for length calculation and 30 fields for crypts number calculation. (C) Snord17^+/+ mice were crossed with Lgr5 ^{GFP‐CreERT2} mice for Lgr5 observation. Representative images of stained jejunum sections are shown. Scale bars, 100 µm. Lgr5⁺ ISC number per filed are quantified and shown in the right panel. n = 30 fields for each group. (D) Small intestine sections of Snord17^+/+ and Snord17 ^−/− mice were stained with Olfm4 antibody. Scale bar, 100 µm. Statistical analysis of Olfm4⁺ cell number per crypt is shown in the right panel. n = 50 crypts for each group. (E) Jejunum sections of Snord17^+/+ and Snord17 ^−/− mice were stained with Ki67 antibody to observe proliferating cells. Typical images are shown in the left panel, and Ki67⁺ cells per crypt are calculated and shown in the right panel. Scale bars, 100 µm. n = 30 crypts for each group. (F) Ki67 staining of organoids from Snord17^+/+ and Snord17 ^−/− mice. Representative images are shown in the left panel, and the percentage of Ki67⁺ cells per organoid are calculated and shown in the right panel. Scale bars, 50 µm. n = 6 organoids for each group. (G) Immunofluorescence staining of cleaved caspase3 in Snord17^+/+ and Snord17 ^−/− intestinal organoids. Scale bars, 50 µm. Percentage of C‐Caspase3⁺ cells per organoid is shown in the right panel. n = 6 organoids for each group. (H) Snord17^+/+ and Snord17 ^−/− mice were treated with 8‐Gy radiation, and small intestines were isolated at day 0, 3, 5 to perform H&E staining. Representative images of jejunum sections are shown. Scale bars, 100 µm. Intact crypt number per filed are calculated. n = 20 fields for each group. (I) Organoid formation assay was performed on crypts from Snord17^+/+ and Snord17 ^−/− mice. Lentivirus was used for Snord17 overexpression in Snord17 ^−/− organoids. Scale bars, 100 µm. n = 5 independent experiments for each group. Data are shown as the means ± SD. Statistical analysis was performed using unpaired two‐tailed Student's t‐tests (A–C,F,G), Mann–Whitney U test (D,E), Mann–Whitney U test with Holm–Šidák correction for multiple comparisons (H), and one‐way analysis of variance (ANOVA) with Tukey's post hoc test (I). ^*** p < 0.001, ns, not significant.

2.3. Snord17 Interacts with Thoc3 in ISCs

Considering canonical functions of snoRNAs, we first validated that whether Snord17 regulated the self‐renewal of ISCs via mediating rRNA modifications. SNOOPY database predicted that Snord17 mediated 2′‐O‐methylation of 28S rRNA at U3474 position (Figure S3A). Given that higher 2'‐O‐methylation levels causes reduced reverse transcription efficiency under low dNTP concentrations [26], we analyzed 2'‐O‐methylation levels of 28S rRNA at U3474 in Snord17^+/+ and Snord17 ^−/− ISCs via qRT‐PCR (Figure S3B). We observed that Snord17 deletion actually reduced 2'‐O‐methylation levels at 28S‐U3474 and reintroduction of wild‐type Snord17 restored it (Figure S3C). However, reintroducing Snord17 with mutated complementary sequences to 28S rRNA failed to rescue the 2'‐O‐methylation level (Figure S3A,C). Interestingly, reintroduction of mutated Snord17 restored impaired organoid formation in Snord17‐deficient ISCs, suggesting that Snord17 participates in the regulation of ISCs in a noncanonical manner (Figure S3D).

We conducted RNA pulldown assays with mass spectrometry to identify Snord17‐interacting protein candidates by using biotin‐labeled Snord17 as a probe. Although mass spectrometry identified three potentially associated proteins, namely Rpl4, macroH2A.1, and Thoc3 (Figure 3A), subsequent Western blot analysis confirmed that Thoc3 specifically interacted with Snord17. In contrast, Rpl4 and macroH2A.1 were detected across the Beads, Sense, and Scramble control groups. (Figure 3B; Figure S4A,B). RNA immunoprecipitation (RIP) with anti‐Thoc3 antibody demonstrated remarkable enrichment of Snord17 compared to IgG controls (Figure 3C). As expected, RNA antisense purification (RAP) assays showed complete loss of Snord17‐Thoc3 interaction in Snord17 ^−/− mice (Figure 3D). In addition, subsequent domain mapping assays identified the △WD6 domain was required for this interaction (Figure 3E). Electrophoretic mobility shift assay (EMSA) further confirmed direct binding between Snord17 and Thoc3 (Figure 3F). Of note, mutated Snord17 retained binding capacity to Thoc3 (Figure S4C). In addition, Snord17 was co‐localized with Thoc3 in ISCs (Figure 3G), and this colocalization was also observed at budding sites of organoids (Figure 3H). We observed that Thoc3 was expressed at higher levels in the crypts compared to the IECs of the small intestine (Figure S4D). Notably, Thoc3 was particularly enriched in Lgr5⁺ stem cells within the crypts, a pattern consistently observed in both the small intestine and colon (Figure S4E,F). However, no obvious difference in Thoc3 expression was detected between Snord17^+/+ and Snord17 ^−/− mice (Figure S4G,H).

FIGURE 3.

Snord17 interacts with Thoc3 in ISCs. (A) Biotin labeled Snord17, scramble sequences and beads were incubated with crypts lysates, then enriched proteins by streptavidin beads were separated via SDS‐PAGE and identified with mass spectrum. Black arrow indicates Thoc3. (B) Immunoblotting analysis of RNA pulldown samples to confirm the interaction between Snord17 and Thoc3. (C) RIP assay was conducted using anti‐Thoc3 antibody or IgG in crypts lysates and Snord17 enrichment was detected by qRT‐PCR. n = 4 independent experiments for each group. (D) The interaction of Snord17 with Thoc3 in Snord17^+/+ and Snord17 ^−/− crypts was examined using RAP assays. (E) Validation of Snord17‐Thoc3 truncation binding specificity via RNA pulldown coupled with anti‐Thoc3 immunoblotting quantification. (F) Recombinant Thoc3 proteins and biotin‐labeled Snord17 were incubated for the EMSA assay. (G) Immunofluorescence analysis of spatial colocalization between Snord17 (green) and Thoc3 (red) in ISCs. Nuclei were stained with DAPI (blue). Scale bar, 10 µm. (H) Snord17 was colocalized with Thoc3 in organoids. Scale bar, 20 µm. (I) Thoc3 knockdown or overexpression was performed on ISCs from Snord17^+/+ and Snord17 ^−/− mice. Representative images of indicated organoids are shown to the left, and the organoid formation ratio per crypt is shown in the right panel. Scale bar, 100 µm. n = 6 independent experiments for each group. (J) Representative H&E staining images of jejunum sections from Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgThoc3, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgThoc3 mice. Scale bars, 100 µm. Length of villus and crypt is shown in the right panel. n = 30 villi and crypts for each group. Data are shown as the means ± SD. Statistical analysis was performed using Welch's t test (C) and one‐way ANOVA with Tukey's post hoc test (I,J). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ns, not significant.

To explore the role of Thoc3 in the regulation of ISC stemness, we performed Thoc3 knockdown or overexpression in Snord17^+/+ and Snord17 ^−/− ISCs, followed by organoid formation (Figure 3I; Figure S4I). We found that Thoc3 knockdown in Snord17^+/+ ISCs significantly impaired organoid formation efficiency, while overexpression of Thoc3 in Snord17 ^−/− ISCs failed to restore organoid formation capacity, suggesting that Snord17 is required for Thoc3‐mediated regulation of ISC stemness (Figure 3I). In addition, Thoc3 knockdown in Snord17^+/+ ISCs inhibited organoid proliferative activity, and overexpression of Thoc3 in Snord17 ^−/− ISCs still suppressed organoid proliferative capacity (Figure S4J). To further investigate the physiological role of Thoc3 in maintaining ISCs stemness in vivo, we generated Thoc3‐deficient mice using CRISPR/Cas9‐mediated genome editing. Thoc3 was successfully deleted in both Snord17^+/+ and Snord17 ^−/− ISCs (Figure S4K). Knockout of Thoc3 caused shortened intestinal crypts and villi in both Snord17^+/+ and Snord17 ^−/− mice (Figure 3J; Figure S4L). Collectively, these data indicate that Snord17 regulates ISCs stemness maintenance via direct interaction with Thoc3.

2.4. Snord17 Facilitates Thoc3‐Dependent Nuclear Export of Yy2 mRNA

We further wanted to explore the molecular mechanism of Snord17‐Thoc3 mediated ISC stemness. As a core subunit of the THO complex, Thoc3 integrates into the TREX complex alongside partners like DEAD box helicase 39b (Ddx39b) and Aly/REF export factor (Alyref) [27, 28], mediating efficient transport of mature transcripts from the nucleus to the cytoplasm, a process critical for fundamental cellular processes [18, 29]. To determine whether Snord17‐Thoc3 regulated ISC stemness by controlling nuclear export efficiency of stemness‐associated transcripts, we performed transcriptome sequencing on total, cytoplasmic, and nuclear components of Snord17^+/+ and Snord17 ^−/− ISCs. Perturbation of mRNA nuclear export generally modulates the subcellular distribution of transcripts without significantly altering their total abundance [30, 31]. Therefore, we performed Gene Ontology (GO) enrichment analysis on genes whose mRNAs that are not altered at the whole cell level (|log₂FC| < 1), while the relative change in cytoplasmic to nuclear (C/N) ratio was ≥ 2. We noticed that these genes were significantly enriched in the regulation of transcription by RNA polymerase II and the regulation of DNA‐templated transcription pathways (Figure 4A), and these genes were predominantly transcription factors (TFs). Further analysis of transcription factor expression identified six TFs with remarkable reduced cytoplasmic/nuclear mRNA ratios but unchanged total mRNA abundance (Figure 4B; Figure S5A). Meanwhile, qRT‐PCR analysis identified Yy2 as the most significantly affected TF, exhibiting a 6‐fold decrease in cytoplasmic/nuclear distribution (Figure 4C). We then explored the impact of the above six TFs on ISC self‐renewal via shRNA and found that Yy2 knockdown significantly impaired ISC self‐renewal capacity (Figure 4D; Figure S5B,C). Furthermore, analysis of Yy2 mRNA distribution via RNA FISH, coupled with mRNA intensity quantification confirmed abolished Yy2 mRNA nuclear export (Figure 4E–G). As expected, Yy2 protein levels were decreased in Snord17 ^−/− ISCs compared to Snord17^+/+ littermates (Figure 4H,I). Intriguingly, we found that Yy2 was predominantly expressed in crypts, exhibiting colocalization with Lgr5⁺ ISCs, indicating its possible function in regulating ISCs stemness (Figure 4J; Figure S5D,E). ​​In addition, knockdown of Yy2 significantly impaired organoid formation capacity and reduced the number of proliferating cells. By contrast, overexpression of Yy2 in Snord17 ^−/− organoids partially rescued organoid formation and increased the proportion of proliferating cells (Figure 4K; Figure S5F,G). Given the established role as an mRNA export mediator of Thoc3, we hypothesized that Snord17 deficiency might impair the binding capacity between Yy2 mRNA and Thoc3. RIP assays demonstrated substantially reduced Yy2 mRNA enriched by Thoc3 antibody in Snord17 ^−/− ISCs (Figure S5H). Of note, Thoc3 knockdown caused reduced Yy2 protein levels, while Thoc3 overexpression in Snord17 ^−/− ISCs failed to restore Yy2 abundance (Figure S5I). Collectively, these results indicate that Snord17 mediates Thoc3‐dependent Yy2 mRNA export.

FIGURE 4.

Snord17 knockout impairs Yy2 mRNA nuclear export. (A) Total, cytoplasmic and nuclear components of Snord17^+/+ and Snord17 ^−/− ISCs were isolated for transcriptome microarray. GO enrichment analysis on genes with comparable total mRNA levels (|log2FC| <1) but exhibited differentiated cytoplasmic‐to‐nuclear mRNA ratio (|log2FC| ≥1). (B) Transcription factor genes exhibiting comparable total mRNA levels were analyzed for cytoplasmic‐to‐nuclear mRNA ratio. Data are shown as a scatter plot. CN: Cytoplasm tpm/Nuclear tpm. (C) Cytoplasmic‐to‐nuclear mRNA level of six TFs (Ybx2, Neurod1, Rarg, Csrnp2, Snpac4 and Yy2) was detected by qRT‐PCR. n = 4 independent experiments for each group. (D) Indicated TFs knockdown was performed on ISCs, followed by organoid formation. Organoid formation ratios are shown. n = 5 independent experiments for each group. (E) Yy2 mRNA localization was visualized by FISH in ISCs from Snord17^+/+ and Snord17 ^−/− mice. Scale bar, 10 µm. (F) Cytoplasmic/Nuclear ratio of Yy2 mRNA (normalized) in ISCs from Snord17^+/+ and Snord17 ^−/− mice is shown. n = 18 cells for each group. (G) Yy2 mRNA intensity analysis was performed using ImageJ. Analyzed axis are indicated by thicker white lines in (E). Labels A and B denote the origin and terminus of the axis, respectively. (H) Yy2 expression in ISCs from Snord17^+/+ and Snord17 ^−/− mice is detected by immunoblotting. (I) Yy2 expression in organoids from Snord17^+/+ and Snord17 ^−/− mice was visualized by immunofluorescence staining. Scale bar, 100 µm. (J) Small intestines isolated from Lgr5 ^{GFP‐CreERT2} mice were stained with Yy2 antibody (red). (K) Yy2 knockdown was conducted in ISCs from Snord17^+/+ mice and Yy2 overexpression was conducted in ISCs from Snord17 ^−/− mice, followed by organoid formation. Representative images of indicated organoids and organoid formation ratios are shown. Scale bar, 100 µm. n = 6 independent experiments for each group. Data are shown as the means ± SD. Statistical analysis was performed using unpaired two‐tailed Student's t‐tests with Holm–Šidák correction for multiple comparisons (C), Kruskal–Wallis test with Dunn's post hoc test (D), Mann–Whitney U test (F), and one‐way ANOVA with Tukey's post hoc test (K). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ns, not significant.

2.5. Yy2 Initiates Tead4 Transcription to Activate Hippo Signaling in ISCs

To identify Yy2 candidate target genes, we conducted transcriptome analysis of Snord17^+/+ and Snord17 ^−/− ISCs. Integrative GSEA and GO analysis revealed that Snord17 deletion profoundly suppressed Hippo signaling, with downregulated downstream genes (Figure 5A; Figure S6A,B). Expression levels of Ccn1 and Ccn2, two canonical Hippo pathway target genes, were reduced post Snord17 deletion (Figure S6C). In contrast, Wnt, BMP, and Notch signaling pathways exhibited no significant alterations (Figure S6D). Among modulators associated with Hippo signaling, Tead4 exhibited the most prominently reduced expression in Snord17 ^−/− ISCs, which was consistent with Yy2 silenced ISCs (Figure 5B,C), suggesting that Yy2 might promote Tead4 transcription. In contrast, expression levels of other core components of the Hippo pathway, including Mst1, Mst2, Lats1, Lats2, Yap1, and Taz, remained largely unchanged following Snord17 knockout (Figure 5B,D). We next performed chromatin immunoprecipitation (ChIP) assays and revealed that Yy2 was enriched on the −1600 to −1400 region of Tead4 promoter (Figure 5E). Subsequent EMSA assay confirmed direct Yy2 binding to this promoter fragment (Figure 5F), which was further verified by luciferase reporter assay (Figure 5G). Given that the trimethylation of histone H3 at lysine 4 (H3K4me3) is associated with an open chromatin state conducive to gene transcription, while the trimethylation of histone H3 at lysine 27 (H3K27me3) is linked to reduced chromatin accessibility and impaired transcriptional activity [32], we assessed histone methylation status at the Tead4 promoter in Snord17 deleted ISCs. Our results demonstrated that Snord17 knockout substantially increased H3K27me3 levels while reducing H3K4me3 modification specifically within the −1600 to −1400 region of the Tead4 promoter (Figure 5H,I). Furthermore, decreased DNase I hypersensitivity at the Tead4 promoter in Snord17 ^−/− ISCs suggested a lower chromatin accessibility and transcriptional activity (Figure 5J). As expected, Tead4 was highly expressed in Lgr5⁺ ISCs compared with Lgr5⁻ IECs (Figure S6E). Depletion of Tead4 significantly impaired organoid formation efficiency and reduced the proportion of proliferating cells. In contrast, ​​overexpression of Tead4 in Snord17 ^−/− organoids effectively restored these impaired phenotypes (Figure 5K; Figure S6F,G).​​ These observations were ​​consistent with corresponding alterations in the expression level of Olfm4​​, a marker of intestinal stem cells, and Ccn1 and Ccn2, canonical Hippo pathway target genes (Figure S6H,I). A ​​parallel change in Tead4 expression​​ was observed in colonic ISCs, ​​implicating a similar functional mechanism​​ between the colon and small intestine (Figure S6J). Taken together, Yy2 promotes Tead4 transcription in ISCs to activate Hippo signaling.

FIGURE 5.

Yy2 enriches on the Tead4 promoter to initiate its transcription for activation of Hippo signaling. (A) GSEA analysis showed enrichment of differentially expressed genes (DEGs) between Snord17^+/+ and Snord17 ^−/− ISCs in the Hippo signaling pathway (NES, normalized enrichment score; FDR, false discovery rate; FWER, family‐wise error rate). (B) Expression levels of genes related to Hippo signaling in Snord17^+/+ and Snord17 ^−/− ISCs were detected using qRT‐PCR. n = 4 independent experiments for each group. (C) Expression levels of Hippo signaling associated genes in Yy2 knockdown and control ISCs were detected using qRT‐PCR. n = 4 independent experiments for each group. (D) Protein levels of core Hippo pathway components were detected in Snord17^+/+ and Snord17 ^−/− ISCs. (E) ChIP combined qPCR analysis of the Tead4 promoter region occupied by Yy2. n = 4 independent experiments for each group. (F) EMSA validation of Yy2 binding to the Tead4 promoter. (G) Luciferase reporter assay was performed to validate Yy2 function on Tead4 transcription activation. FL, full‐length (represent −2000∼0 region upstream Tead4 transcriptional start site). Δ‐1600∼‐1400 (represent −2000∼0 region without ‐1600∼‐1400 region upstream Tead4 transcriptional start site). n = 6 independent experiments for each group. (H) Enrichment of H3K27me3 on Tead4 promoter in Snord17^+/+ and Snord17 ^−/− ISCs was analyzed by ChIP assay. n = 4 independent experiments for each group. (I) Enrichment of H3K4me3 on Tead4 promoter in Snord17^+/+ and Snord17 ^−/− ISCs was analyzed by ChIP assay. n = 4 independent experiments for each group. (J) Chromatin accessibility of the Tead4 promoter was detected by DNase I digestion assays. n = 6 independent experiments for each group. (K) Organoid formation was conducted with Tead4 knockdown in Snord17^+/+ ISCs and Tead4 overexpression in Snord17 ^−/− ISCs. Typical images of indicated organoids are shown in the left panel and organoid formation ratios are shown in the right panel. Scale bar, 100 µm. n = 6 independent experiments for each group. Data are shown as the means ± SD. Statistical analysis was performed using unpaired two‐tailed Student's t‐tests with Holm–Šidák correction for multiple comparisons (B,C,E,H), Welch's ANOVA with Dunnett T3's post hoc test (G), unpaired two‐tailed Student's t‐tests (I,J) and one‐way ANOVA with Tukey's post hoc test (K). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001.

2.6. Tead4 Deletion Impairs Self‐Renewal Maintenance of ISCs and Intestinal Regeneration

To test the physiological role of Tead4 in maintaining ISC stemness, we generated Tead4‐deficient mice via CRISPR/Cas9‐mediated genome editing. Tead4 was deleted in Snord17^+/+ and Snord17 ^−/− ISCs (Figure S7A). Tead4 knockout caused shortened intestinal crypts and villi in both Snord17^+/+ and Snord17 ^−/− mice (Figure 6A; Figure S7B,C). Moreover, colon crypts exhibited decreased length in Tead4‐deficient mice (Figure S7D,E). In addition, Tead4 deletion significantly reduced Ki67⁺ proliferating cells in the small intestines (Figure 6B). We then conducted organoid formation assays using crypts isolated from Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgTead4, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgTead4 mice. We found that Tead4 deficiency markedly impaired organoid formation and reduced Ki67⁺ proliferating cells (Figure 6C,D). To assess intestinal regeneration in Tead4‐deficient mice, we subjected mice to irradiation (8 Gy). Tead4 deletion severely impaired epithelial injury repair in both Snord17^+/+ and Snord17 ^−/− mice (Figure 6E). Collectively, these results indicate that Tead4 promotes self‐renewal of ISCs and intestinal regeneration.​

FIGURE 6.

Tead4 deletion disrupts self‐renewal capacity of ISCs. (A) Representative H&E staining images of jejunum sections from Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgTead4, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgTead4 mice. Scale bars, 100 µm. Length of villus and crypt are shown in the right panel. n = 30 villi and crypts for each group. (B) Small intestine jejunum sections of Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgTead4, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgTead4 mice were stained with Ki67 antibody to observe proliferating cells. Typical images are shown in the left panel, and Ki67⁺ cells per crypt are shown in the right panel. Scale bars, 100 µm. n = 30 crypts for each group. (C) Organoid formation assay was performed on crypts from Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgTead4, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgTead4 mice. Representative images of indicated organoids are shown, and the organoid formation ratio per crypt is shown in the lower panel. Scale bar, 50 µm. n = 8 independent experiments for each group. (D) Ki67 staining of organoids from Snord17^+/+ ;sgCtrl, Snord17^+/+ ;sgTead4, Snord17 ^−/−;sgCtrl, and Snord17 ^−/−;sgTead4 mice. Representative images are shown in the left panel, and the percentage of Ki67⁺ cells per organoid is shown in the right panel. Scale bars, 50 µm. n = 8 organoids for each group. (E) H&E staining of intestinal tissues from indicated mice at day 0, 3, 5 post radiation. H&E staining of the jejunum sections is shown as representatives. Intact crypt number per field is shown in the right panel. Scale bars, 100 µm. n = 20 fields for each group. Data are shown as the means ± SD. Statistical analysis was performed using one‐way ANOVA with Tukey's post hoc test (A,C,D) and Kruskal–Wallis test with Dunn's post hoc test (B,E). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ns, not significant.

Given the critical role of ISCs in orchestrating intestinal injury repair during colitis and driving the pathogenesis of colorectal cancer (CRC), we further explored the involvement of SNORD17 in IBD and CRC patients [33, 34, 35]. Analysis of clinical samples from patients with IBD revealed that expression levels of SNORD17 were elevated in regions with obvious inflammation compared to non‐inflamed regions (Figure S8A), suggesting that the inflammatory milieu may actively stimulate stemness potential of ISCs, possibly as an adaptive mechanism to initiate tissue regeneration. In addition, an analysis of the TCGA database indicated that SNORD17 expression was significantly higher in patients with CRC than in normal individuals (Figure S8B), which was consistent with the results of qRT‐PCR assays performed on our subsequent clinical samples (Figure S8C). Bioinformatics analysis of TCGA data also demonstrated elevated expression levels of THOC3, YY2, and TEAD4 in CRC samples compared to normal controls (Figure S8D–F). Furthermore, knockdown of either SNORD17 or TEAD4 in CRC patient‐derived organoids resulted in a significant reduction in the size and growth of organoids (Figure S8G,H), indicating a disruption in the self‐renewal and proliferative capacity.

3. Discussion

Despite the complicated​ regulatory networks that govern ISC self‐renewal and fate decisions remain incompletely elucidated [36], the core signaling pathways, such as Wnt, Notch, and BMP are considered as key players. The full spectrum of molecular regulators, especially epigenetic and post‐transcriptional modulators, has not been defined yet [34]. Our study reveals a critical and previously unrecognized role for the small nucleolar RNA Snord17 in regulating the self‐renewal capacity of ISCs. Snord17, which is highly expressed in ISCs, physically associates with the mRNA export regulator Thoc3 to promote the nuclear export of Yy2 mRNA. This post‐transcriptional regulatory step is essential for maintaining adequate Yy2 protein levels. The synthesized Yy2 then translocates back into the nucleus, where it functions as a transcriptional activator of Tead4, thereby initiating the Hippo signaling pathway and ultimately enhancing ISC self‐renewal (Figure S9). Importantly, this SNORD17‐THOC3‐YY2‐TEAD4 regulatory axis is aberrantly regulated in both IBD and CRC patients, implicating its potential contribution to disease pathogenesis.

In conventional dogma, snoRNAs have been acknowledged for their well‐established functions in guiding 2′‐O‐methylation and pseudouridylation modifications of rRNAs. However, a growing body of evidence in recent years has expanded their functions to regulate diverse cellular processes beyond the classical roles [13, 14, 37]. Despite this accumulating evidence of non‐canonical functions, their roles in regulating self‐renewal capacity of ISCs remain almost entirely unexplored. In this study, we identified a snoRNA Snord17 that is highly expressed in ISCs and further demonstrated that Snord17 is required for maintaining ISC self‐renewal. Snord17‐mediated rRNA modification is not involved in the regulation of ISC stemness. Of note, we showed a novel mechanism in which Snord17 facilitates the nuclear export of Yy2 mRNA, a post‐transcriptional regulatory function, which has not been previously characterized in any other snoRNA.

Gene expression in cells is often regulated by a cellular microenvironment [38, 39]. ISCs reside within a complex niche at the bottom of crypt and are subjected to multifaceted regulation via various extrinsic signals. For example, neighboring Paneth cells are known to secrete factors such as Wnt3a and Dll4 that critically influence ISC behavior [40]. Here, we showed that Snord17 is highly expressed in ISCs, and its expression is dysregulated in inflammatory and cancerous environments. Whether the expression of Snord17 in ISCs under different conditions is modulated by niche‐derived signals, such as those from Paneth cells or other stromal components, is an important and interesting topic that requires further investigation.

The nuclear export of mRNA is a highly complicated process that requires the TREX complex to bind nascent transcripts and recruit an export receptor NXF1‐NXT1, facilitating mRNA translocation through a nuclear pore complex [28]. Thoc3, a core component of the TREX complex, interacts with the small nucleolar RNA Snord17, which revealed a specific regulatory mechanism. Although Snord17 deletion does not alter the expression level of Thoc3, it dramatically impaired the binding affinity between Thoc3 and Yy2 mRNA, thereby disrupting the nuclear export of Yy2 transcripts. These findings indicate that Snord17 potentially acts as a molecular bridge between Thoc3 and Yy2 mRNA. The precise machinery of this facilitation, how it involves guiding Thoc3 to the specific mRNA, stabilizing the interaction, or conferring specificity, represents a compelling and unresolved question that merits further investigation.

The Hippo pathway integrates a diverse array of extracellular and intracellular signals, including cell contact inhibition, mechanical stimuli, G‐protein‐coupled receptor (GPCR) signaling, energy, and metabolic stress [21]. Recent study also indicates that the epigenetic regulator CBX4 influences genomic instability, thereby modulating Hippo pathway activity [41]. Similarly, the increase in H3K27me3 and concomitant decrease in H3K4me3 at the Tead4 promoter upon Snord17 depletion implies a potential role in modulating a local chromatin landscape. This intriguing observation opens the possibility that the Snord17‐Thoc3 axis may also involve epigenetic fine‐tuning to ensure the correctness of the transcriptional program essential for ISC stemness maintenance.

4. Materials and Methods

4.1. Antibodies and Reagents

Anti‐Ki67 (Cat# ab15580), Anti‐nucleolin (Cat# ab129200) and F‐actin staining kit (Cat# AB1112127) were all obtained from Abcam. Anti‐H3 (Cat#4499), Anti‐H3K27me3 (Cat# 9733), Anti‐Olfm4 (Cat# 39141S), YAP (D8H1X) XP rabbit monoclonal antibody (Cat# 14074), and Anti‐Cleaved Caspase‐3 (Cat# 9661T) were purchased from Cell Signaling Technology. Anti‐EEA1 was purchased from Santa Cruz. Anti‐Tead4 (Cat# 12418‐1‐AP), MST1 polyclonal antibody (Cat# 22245‐1‐AP), LATS2‐specific polyclonal antibody (Cat# 20276‐1‐AP) and Ribosomal protein L4 polyclonal antibody (Cat# 11302‐1‐AP) were purchased from Proteintech. Anti‐Thoc3 (Cat# NBP1‐92503) and Anti‐Yy2 (NBP2‐93905) were purchased from Novus. Anti‐Flag (Cat# F1804) antibodies were from Sigma‐Aldrich. Anti‐EpCAM (Cat# 118212) and Anti‐GFP (Cat# 338001) were from Biolegend. Anti‐β‐actin (Cat# RM2001) and HRP‐conjugated secondary antibodies were purchased from Beijing Ray Antibody Biotech. Alexa‐488, Alexa‐594, and Alexa‐647 conjugated anti‐rabbit and anti‐mouse secondary antibodies were purchased from Invitrogen. macroH2A.1 (2T18) rabbit monoclonal antibody (Cat# RM5533) and WWTR1 rabbit polyclonal antibody (C‐term) (Cat# BD‐PB4907) were obtained from Biodragon. Anti‐MST2 rabbit monoclonal antibody (Cat# R014203) was purchased from Epizyme. LATS1 rabbit polyclonal antibody (Cat# 252567) and SNX5 rabbit monoclonal antibody (Cat# R22669) were purchased from Zenbio. Biotin RNA Labeling Mix (Cat# 11685597910), DIG RNA Labeling Mix (Cat# 11277073910), and T7 RNA polymerase (Cat# 10881767001) were purchased from Roche. M‐MLV Reverse Transcriptase (Cat# M1701) was from Promega. Dual Luciferase Reporter Gene Assay Kit (Cat# RG027), Glucose Assay Kit with O‐toluidine (Cat# S0201S), and Amplex Red Free Fatty Acid Assay Kit (Cat# S0215S) were from Beyotime. Paraformaldehyde (PFA) and 4’,6‐diamidino‐2‐phenylindole (DAPI) were from Sigma.

4.2. Generation of Knockout Mice by CRISPR‐Cas9 Technology

Snord17 ^−/−mice were generated using CRISPR‐Cas9 method as described previously [42]. About 250 zygotes from C57BL/6 background mice were injected with corresponding CRISPR‐mediated single‐stranded guide RNA (sgRNA) donors (Table S1) and subsequently transferred to the uterus of pseudo‐pregnant females from which viable founder mice were obtained. The DNA from founder mice was subjected to PCR amplification, followed by sequencing, Northern blotting, and qRT‐PCR to validate the knockout of Snord17. The expression of host gene Snx5 was also analyzed via qRT‐PCR and Western blotting. The following primers were used for PCR amplification of Snord17: 5′‐ TGCAGGAAGAGGGCATCAA‐3′ and 5′‐TTGGGCAACTTCTGTGTTACT‐3′.

Cas9‐KI and Lgr5‐EGFP‐IRES‐CreERT2 (Lgr5 ^{GFP‐CreERT2}) mice were purchased from the Jackson Laboratory. All the mice were C57BL/6 background and 6‐8 weeks old. We used littermates with the same age and gender for each group. Animal use and protocols were approved by the Institutional Animal Care and Use Committees (approval number: SYXK2024221) at the Institute of Biophysics, Chinese Academy of Sciences.

4.3. RNA Interference

Silencing of indicted genes was performed via short hairpin RNA as described before [43]. shRNA was designed on an online shRNA designer (https://biosettia.com/support/shrna‐designer/). Target sequences are listed in Table S2. shRNA was constructed into pSicor‐Puro lentivirus vector and co‐transfected with packaging plasmids psPAX2 (Cat# 12260, Addgene) and pMD2G (Cat# 12259, Addgene) into HEK293T cells for virus production by using Lipofectamine 3000. Viral particle enrichment was performed by sequential collection of culture supernatants at 24 h intervals (days 1‐3 post‐transfection), filtered through 0.45 µm filters to remove cellular debris, followed by ultracentrifugation at 25 000 rpm for 2 h at 4°C for viral concentration. Organoids were infected with collected virus and selected by puromycin.

4.4. Crypt Isolation and Organoid Culture

Intestinal crypts isolation and organoids culture were performed as previously described with minor modifications [44]. Intestines were isolated from sacrificed mice and opened longitudinally followed by pre‐cooled PBS washing. Intestinal tissues were dissected into 1 cm fragments and enzymatically digested in pre‐warmed collagenase solution (DMEM/F12 medium containing 0.1% type I collagenase (Invitrogen), 100 units/mL penicillin, 0.1 mg/mL streptomycin, and 10 mm HEPES) at 37°C for 24 min. Every 8 min, intestine fragments were dissociated through pipetting, and mixtures were checked with microscopy until a high percentage of intact crypts was observed. Then the mixtures were collected and filtered through a 70 µm cell strainer, followed by centrifugation at 80 g for 2 min. Collected crypts were resuspended with organoids culture medium, mixed with Matrigel, and seeded in 24‐well plate. The plates were incubated in a culture incubator until Matrigel polymerizes and prewarmed organoids culture medium was added for organoids formation and refreshed every two days.

For organoids passaging, the organoids embedded in Matrigel in each well were resuspended with cool DMEM/F12 medium, broken into fragments through pipetting, and spun at 600 g for 5 min. The pellets were resuspended with fresh cool organoids culture medium, mixed with Matrigel, and seeded in 24‐well plate as described above.

4.5. Quantitative Real‐Time PCR

Total RNA from cells or tissues was extracted using Trizol. The quantity and quality of RNA were determined by NanoDrop (Thermo Fisher). Reverse transcription was performed using the 5×All‐In‐one RT Mastermix (Abm, Vancouver, Canada), and real‐time PCR was conducted with specific primers. Primers are listed in Table S3.

4.6. Northern Blotting

Total RNA was extracted from different tissues with TRIzol and separated via denaturing urea‐polyacrylamide gel electrophoresis in 0.5×TBE buffer for 2 h. RNA was transferred to a positively charged nylon membrane in 0.5×TBE buffer and UV‐crosslinked at 240 mJ/cm² (265 nm). Membranes were pre‐hybridized and hybridized with biotin‐labeled RNA probes (65°C, 16–20 h). After washing, signals were detected using a Chemiluminescent Nucleic Acid Detection Kit (Thermo Fisher Scientific, #89880) following the manufacturer's instruction.

4.7. Biotin‐Labeled RNA Pulldown and Mass Spectrometry Assay

Biotin‐labeled RNAs were synthesized by in vitro transcription using T7 RNA polymerase (Roche) with Biotin RNA Labeling Mix (Roche). Following isolation, intestinal crypts were lysed in ice‐cold RIPA buffer (GenStar) at 4°C for 1 h. Lysates were incubated with probes overnight at 4°C, followed by the 4 h incubation with streptavidin‐agarose beads to enrich RNA‐binding proteins. Captured proteins were resolved via SDS‐PAGE and detected by silver staining. Protein bands specifically enriched by biotin‐Snord17 were excised for analysis using a Q Exactive mass spectrometer (Thermo Fisher Scientific).

4.8. Fluorescence In Situ Hybridization

For Snord17 and Yy2 in situ hybridization, intestinal sections or ISCs were fixed in 4% PFA, permeabilized with 1% Triton X‐100 (30 min), and hybridized with biotinylated probes in FISH buffer (50% formamide, 2× SSC, 0.5 mg/mL yeast tRNA, 0.5 mg/mL salmon sperm DNA, 2.5 mg/mL BSA) at 55°C for 1 h. SSC buffer was used for post‐hybridization washing. Signals were amplified with the Fluorescent in Situ Hybridization Kit (RiboBio). The sections were mounted in Fluoromount‐G (Southern Biotech) and imaged on a Nikon A1R+ confocal microscope.

4.9. Immunofluorescence Staining

For tissue section staining, small intestines and colons were longitudinally dissected, PBS‐rinsed, and fixed in 4% paraformaldehyde (1 h). Samples were dehydrated in 30% sucrose (24 h) prior to OCT embedding. Sectioned tissues underwent PBS rehydration, 1% Triton X‐100 permeabilization (in PBS, 1 h), and blocking with 10% donkey serum in PBS (30 min). Primary antibody incubation (4°C, overnight) was performed, followed by species‐matched Alexa Fluor‐conjugated secondary antibodies incubation (room temperature, 1 h). Mounted slides were imaged using a confocal microscope (Nikon A1R+) and analyzed with Imaris 9. We performed immunofluorescence staining to three intestinal segments (duodenum, jejunum, and ileum) and presented immunofluorescence staining images of the jejunum as a representative image in the figures. For Lgr5⁺ cell number calculation, each Lgr5‐GFP⁺ signal that was co‐localized with DAPI and displayed characteristic columnar shape was counted as one cell.

4.10. Western Blotting

ISCs, crypts, or organoids were lysed in ice‐cold RIPA lysis buffer (strong) on ice for 30 min. Lysates were centrifugated at 12 000 × g for 10 min at 4°C to remove cellular debris. Protein supernatants were mixed with SDS loading buffer, denatured at 100°C for 30 min, separated via SDS‐PAGE electrophoresis, and then transferred to nitrocellulose membranes (Bio‐Rad). Membranes were blocked with 8% milk in TBST (Tris‐buffered saline with 0.1% Tween‐20) for 1 h at room temperature before incubation with specific primary antibodies diluted in blocking buffer (4°C, overnight). After three 10‐min TBST washes, membranes were incubated with horseradish peroxidase (HRP)‐conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using enhanced chemiluminescence substrate (Thermo Scientific).

4.11. Nucleoplasmic, Cytoplasmic, and Nucleolus Fractionation

Nucleoplasmic, cytoplasmic, and nucleoli fractions were isolated from intestinal crypts using a modified protocol based on prior methods [45]. Briefly, crypts were dissociated into single cells, washed with ice‐cold PBS, and incubated in NSB buffer (10 mm Tris‐Cl pH 7.4, 10 mM NaCl, and 0.5–2 mM MgCl₂) at 4°C for 30 min. After adding 10% NP‐40 to reach 0.3% final concentration, cells were homogenized in a 15 mL Dounce homogenizer with 15 pestle strokes until microscopic examination confirmed intact nuclei devoid of plasma membrane remnants. The homogenate was centrifuged at 1200 ×g for 5 min at 4°C to collect the cytoplasmic supernatant. Nuclear pellets were resuspended in 250 mM sucrose with 10 mM MgCl₂, underlaid with 880 mM sucrose containing 5 mM MgCl₂, and centrifuged at 1200 ×g for 10 min at 4°C. The resulting pellet was resuspended in 340 mM sucrose with 5 mM MgCl₂, sonicated using 8 cycles of 10 s pulses at 400 V with 10 s intervals, layered over 880 mM sucrose, and centrifuged at 2000 ×g for 20 min at 4°C to separate nucleoplasm as supernatant from nucleoli as pellet.

4.12. Electrophoretic Mobility Shift Assay (EMSA)

Biotin‐ or DIG‐labeled Snord17 was synthesized by in vitro transcription using T7 RNA polymerase with respective labeling mixes. Recombinant Thoc3 protein was incubated with Snord17 in EMSA binding buffer and subjected to electrophoretic mobility shift assays using 6% native polyacrylamide gels. Nucleic acid‐protein complexes were transferred to nylon membranes, and biotin signals were detected using the Chemiluminescent Nucleic Acid Detection Kit (Thermo Fisher, 89880).

4.13. Immunohistochemistry Assay

Paraffin‐embedded intestinal sections were deparaffinized through xylene gradients and rehydrated through an ethanol series. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide for 10 min. Antigen retrieval was performed in Tris‐EDTA buffer at 100°C for 15 min. Sections were blocked with 10% donkey serum for 30 min at room temperature before incubation with primary antibodies for 2 h at room temperature. HRP‐polymer‐conjugated secondary antibodies were applied for 1 h, followed by DAB chromogenic development and hematoxylin counterstaining. Slides were dehydrated through graded alcohols and xylene, then mounted in resinous medium for brightfield microscopy.

4.14. Luciferase Reporter Assay

The full‐length promoter of Tead4 (spanning from −2000 to 0 upstream of the transcriptional start site) or a truncated Tead4 promoter variant (Δ‐1600 to ‐1400) was cloned into the pGL3‐basic luciferase reporter vector. The coding sequence (CDS) region of Yy2 was inserted into the pcDNA4 vector for overexpression. One day prior to transfection, HEK293T cells were seeded in 24‐well plates. For each well, a transfection mixture containing 100 ng of the pGL3 reporter plasmid, 1 ng of pRL‐TK (an internal control expressing Renillaluciferase), and 500 ng of the pcDNA4‐Yy2 overexpression plasmid was prepared and delivered into the cells. Luminescence signals were measured using a Dual Luciferase Reporter Gene Assay Kit (Beyotime) following the manufacturer's instructions.

4.15. Flow Cytometry Sorting

Crypts and epithelial cells were isolated from 8 weeks old Lgr5 ^{GFP‐CreERT2} mice as described above. Single‐cell suspensions were obtained by dissociating the crypts with TrypLE™ Express (Thermo Fisher Scientific). Fluorescence‐activated cell sorting (FACS) was then performed to isolate GFP‐positive Lgr5⁺ ISCs from both the small intestine and colon, along with GFP‐negative IECs. RNA was extracted, reverse‐transcribed, and subjected to quantitative analysis of gene expression levels.​

4.16. Analysis of 2′‐O‐methylation at 28S‐U3474

2′‐O‐methylation at 28S rRNA U3474 was analyzed as described before [26]. cDNA synthesis was performed in reactions containing RNase H minus Moloney leukemia virus reverse transcriptase, RNasin ribonuclease inhibitor, and 1 mM of reverse primer targeting the sequence downstream to 28S‐U3474, with either 10 µM or 1 mM dNTPs. Reverse transcription proceeded at 37°C for 5 min followed by enzyme inactivation at 70°C for 15 min. Quantitative PCR amplification was performed and 2′‐O‐methylation levels were calculated as 2^{(CT‐low/CT‐high)}.

4.17. DNase I Sensibility Assay

Nuclei were isolated using a commercial nuclei isolation kit (Cat# 78833). Isolated nuclei were resuspended in 200 µL DNase I digestion buffer and treated with 2 U DNase I at 37°C for 5 min. Total DNA was purified post‐digestion and analyzed by qRT‐PCR.

4.18. IBD and CRC samples

All IBD and CRC patients are from Beijing Friendship Hospital, Capital Medical University. IBD patients were treatment‐naïve and enrolled at the time of diagnostic colonoscopy. For each patient with IBD, biopsy samples were collected from two distinct regions of the ileum: the endoscopically most severely affected (inflamed) area and the endoscopically least affected (non‐inflamed) area. Primary human CRC samples were collected from patients with a pathological diagnosis of colorectal cancer in Beijing Friendship Hospital, Capital Medical University. This study was approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University.

4.19. SnoRNA and Transcriptome Sequencing

SnoRNA profiling was performed in GFP⁺ ISCs and GFP⁻ epithelial cells (IECs) isolated from 8‐week‐old Lgr5 ^{GFP‐CreERT2} mice by flow cytometry. TRIzol‐extracted RNA was analyzed using the Mouse snoRNA Expression Microarray (YingBioTech). For identification of genes regulated by Snord17, RNA from sorted Snord17⁺^/⁺ and Snord17 ^−/− Lgr5^GFP ISCs underwent transcriptome sequencing (BGISEQ platform, BGI Genomics).

4.20. Chromatin Immunoprecipitation (ChIP) Assay

ISCs were crosslinked with 1% formaldehyde at 37°C for 10 min, washed twice in PBS, and lysed with SDS buffer. Chromatin was sonicated to generate 200–500 bp DNA fragments. Lysates were precleared with 50% Protein A Agarose/Salmon Sperm DNA slurry, then incubated with 2 µg specific antibodies overnight at 4°C. Following immunoprecipitation and washes, DNA was eluted using 1% SDS and 0.1 M NaHCO₃, extracted using isopropanol and phenol‐chloroform, and analyzed by qRT‐PCR with primers detailed in Table S4.

4.21. Statistics and Reproducibility

Sample sizes, experimental replicates, and statistical tests are detailed in the Figure legends. The statistical analysis of all datasets, which featured sample sizes below 50 (classified as small sample studies), was performed using ​​GraphPad Prism 10​​. Normality for all data was assessed using the Shapiro‐Wilk test. Homogeneity of variances was verified employing an F‐test for comparisons between two independent groups and Bartlett's test for comparisons involving three or more groups. For comparisons between two independent groups: if data satisfied both normality and variance homogeneity assumptions, a two‐tailed unpaired Student's t‐test was applied; if normality was met but variances were unequal, Welch's t‐test was used; if the data violated the normality assumption, the non‐parametric Mann–Whitney U test was employed. For comparisons among three or more independent groups: if both normality and variance homogeneity were satisfied, one‐way ANOVA was performed, followed by Tukey's post hoc test for multiple comparisons; if normality was met but variance homogeneity was violated, Welch's ANOVA was utilized, followed by Dunnett T3's post hoc test; if the data distribution was non‐normal, the non‐parametric Kruskal–Wallis test was conducted, followed by Dunn's post hoc test. For multiple comparisons correction in pairwise tests, the Holm–Šidák method was applied to control the family‐wise error rate. Results are shown as mean ± SD. p values < 0.05 were considered significant (^* p < 0.05; ^** p < 0.01; ^*** p < 0.001) and p > 0.05 was considered non‐significant (ns).

Author Contributions

P.Z. designed the project, performed experiments, analyzed data, and wrote the paper. Y.X., Y.L., Z.X., and Z.Y. performed experiments. Y.X., Z.X., Z.Y., R.W., C.L., Y.D., and H.G. analyzed data. Y.Y. provided samples and analyzed data. Z.F. initiated the study, organized, designed, and wrote the paper.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File: advs73437‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.