A novel tRF-Gly is associated with obesity development through post-transcriptional regulation of lipid metabolism Authors: Yuhang Lei, Mailin Gan, Kai Wang, Tianci Liao, Yiting Yang, Xue Zhao, Xin Zhang, Dujun Chen, Xinyi Wang, Jianfeng Ma, Lili Niu, Ye Zhao, Lei Chen, Xiaofeng Zhou, Yan Wang, Mingzhou Li, Li Zhu, Linyuan Shen Published: 2025 ABSTRACT AbstractBackgroundObesity, characterized by excessive fat accumulation, represents a global health crisis closely linked to metabolic disorders such as type 2 diabetes, hypertension, and atherosclerosis. tRNA-derived small RNAs (tsRNAs) have recently emerged as important epigenetic regulators, yet their roles in fat deposition remain poorly characterized. This study aims to identify tsRNAs that influence fat accumulation and to elucidate their molecular mechanisms, with a focus on tRF‑Gly‑GCC‑037 (tRF‑Gly) as a candidate regulator of adipocyte differentiation.MethodsVisceral adipose tissue was collected from obese and lean pigs for comprehensive tRF and tiRNA sequencing. Differential expression analysis identified tRF‑Gly as a highly abundant candidate in obese samples. Functional assays in 3T3‑L1 preadipocytes included both overexpression and knockdown of tRF‑Gly, followed by lipid accumulation measurements and assessment of key adipogenic markers (CEBPα and PPARγ) by quantitative real-time PCR (qRT‑PCR) and western blot. Mechanistically, dual‑luciferase reporter assays, RNA immunoprecipitation (RIP), and nuclear–cytoplasmic protein fractionation were performed to examine how tRF‑Gly modulates the RAC1/JNK2/β‑catenin signaling axis.ResultstRF‑Gly was significantly upregulated in visceral adipose tissue from obese pigs and ranked among the most abundant tsRNAs. Overexpression of tRF‑Gly in 3T3‑L1 cells and in C57BL/6 mice promoted lipid accumulation and increased CEBPα and PPARγ expression, whereas tRF‑Gly knockdown reduced lipid deposition. Mechanistically, tRF-Gly was suggested to bind RAC1 mRNA with AGO3 involvement, leading to RAC1 silencing. Consistently, RAC1 knockdown phenocopied the adipogenic effects of tRF-Gly, whereas RAC1 overexpression reversed these effects. Furthermore, RAC1 deficiency disrupted the RAC1/JNK2/β‑catenin complex, impaired β‑catenin nuclear translocation, and suppressed Wnt/β‑catenin signaling.ConclusionsOur findings demonstrate that tRF‑Gly functions as a key regulator of fat accumulation. By silencing RAC1 via AGO3, tRF‑Gly disrupts RAC1/JNK2/β‑catenin complex assembly, prevents β‑catenin nuclear translocation, and downregulates Wnt/β‑catenin signaling, thereby promoting lipid deposition. This study uncovers a novel epigenetic mechanism by which tRF‑Gly controls fat accumulation and suggests that targeting tRF‑Gly may represent a therapeutic strategy for obesity and related metabolic disorders.Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s11658-025-00835-2.Keywords: tRF-Gly, RAC1, RAC1/JNK2/β-catenin transport complex, Wnt/β-catenin signaling pathway, Fat deposition Introduction Obesity and associated metabolic disorders—such as type 2 diabetes, hypertension, hyperlipidemia, and atherosclerosis—pose a serious threat to global health. By 2035, more than 4 billion people (approximately 51% of the world’s population) are projected to be overweight or obese [1]. Obesity arises from a chronic imbalance between energy intake and expenditure, leading to excessive accumulation of white adipose tissue (WAT) [2]. WAT serves as the body’s primary energy reservoir by converting surplus calories into stored triglycerides; excessive WAT expansion underlies the pathology of obesity. Notably, visceral adipose tissue (VAT) enlargement is more strongly associated with obesity-related complications and increased mortality than subcutaneous adipose tissue (SAT) accumulation [3]. Domestic pigs (Sus scrofa) are considered a reliable model for obesity research because of their close anatomical, physiological, and metabolic resemblance to humans [4]. In particular, the distribution and function of adipose depots in pigs closely resemble those in humans, making pigs highly valuable for translational research in fat deposition and metabolic disease [5]. Non-coding RNAs (ncRNAs) are crucial regulators of gene expression at transcriptional, post-transcriptional, and translational levels. Extensive studies have shed light on the underlying mechanisms of these regulatory RNAs, including microRNAs (miRNAs) [6], long non-coding RNAs (lncRNAs) [7], and circular RNAs (circRNAs) [8], influence fat deposition. Notably, tsRNAs, including tRNA-derived fragments (tRFs) and tRNA halves (tiRNAs), have become a key research focus across a wide range of physiological and pathological processes [9]. For example, the 3′tRNA-derived fragment tRF-Val has been shown to directly bind to the chaperone molecule eukaryotic translation elongation factor 1 alpha 1 (EEF1A1), thereby facilitating its nuclear transport and promoting its interaction with MDM2 proto-oncogene (MDM2), a specific p53 E3 ubiquitin ligase. This interaction inhibits the downstream p53 molecular pathway, thereby promoting the onset of gastric cancer [10]. Furthermore, research has revealed that 5′tiRNA-His-GTG responds to hypoxia via the HIF1α/ANG axis, contributing to the development of colorectal cancer through its interaction with large tumor suppressor kinase 2 (LATS2) [11]. In contrast, the regulatory mechanisms of tsRNAs in fat deposition remain less explored, with only a limited number of omics studies available for reference [12–14]. Therefore, to advance our understanding of tsRNA function in obesity, it is essential to dissect the molecular mechanisms by which specific tsRNAs influence lipid storage. In the past two decades, the research on Wnt/β-catenin signaling and its role in fat metabolism have advanced rapidly [15]. Studies have shown that Wnt/β-catenin signaling suppresses adipogenesis by disrupting the transcription of crucial adipogenic regulators, such as CEBPα and PPARγ [16, 17]. The cascade response of the Wnt/β-catenin pathway depends on the nuclear import of the transcriptional regulator β-catenin [18]. When Wnt signaling is inactive, the β-catenin degradation complex effectively degrades β-catenin, limiting its nuclear accumulation and thereby suppressing Wnt/β-catenin pathway activation [19, 20]. The degradation complex primarily consists of glycogen synthase kinase 3 beta (GSK3β), casein kinase 1 alpha 1 (CK1α), and AXIN proteins. AXIN serves as the key scaffold for the complex’s destruction, while CK1α and GSK3β phosphorylate β-catenin at specific sites, including Thr41 and Ser33/37, thereby accelerating its degradation [21, 22]. Dysfunction of the degradation complex impacts both adipocyte differentiation and adipogenesis. For instance, increasing AXIN1 levels to suppress this signaling cascade can expedite adipogenic differentiation [23]. Endogenous inhibition of GSK3β hinders 3T3-L1 preadipocyte maturation. Conversely, exogenous restoration of GSK3β expression facilitates the maturation process in 3T3-L1 preadipocytes [24]. Consistently, recent work in Yanbian cattle demonstrated that Wnt/GSK3/β-catenin signaling in fibro/adipogenic progenitors regulates intramuscular fat deposition, further underscoring the anti-adipogenic role of GSK3 inhibition [25]. Currently, no studies have directly implicated CK1α in the fat deposition process; instead, it is more directly associated with glucose homeostasis [26]. The rapid increase in global obesity and metabolic syndrome underscores the urgent need to better understand molecular drivers of fat deposition and to develop effective therapeutic strategies. In this study, we leveraged extensive tsRNA sequencing data from previous work and observed strong correlations between tsRNA profiles and lipid metabolism [13, 14, 27]. Given the ambiguity surrounding tsRNAs’ molecular roles in fat deposition, our research sought to elucidate their contribution and underlying regulatory processes in fat accumulation. Specifically, we screened and identified tRF-Gly, a differentially expressed and highly abundant tsRNA, from the visceral fat of obese and lean pigs, and demonstrated its significant impact on both adipogenic differentiation in 3T3-L1 preadipocyte and fat deposition in C57BL/6 mice. Mechanistically, we demonstrated that tRF-Gly silences its target gene RAC1 via the AGO3 protein, disrupts RAC1/JNK2/β-catenin complex assembly, limits β-catenin nuclear translocation, suppresses Wnt/β-catenin signaling activity, and ultimately facilitates fat accumulation. By elucidating this novel tsRNA-mediated pathway, our study provides fresh insights into the epigenetic control of fat deposition and identifies tRF-Gly as a potential target for mitigating obesity and its complications. Materials and methods Experimental animals and sample collection No new experimental animals (pigs) were slaughtered in this experiment, and all tissue samples and sequencing data were obtained from our previous study [14, 27]. The experimental animals in this study were obtained from three Qing Yu pigs (QYP) and three Yorkshire pigs (YP) from our previous studies. The pigs were handled in compliance with the stipulations outlined in the Code of Practice for the Quality Management of Livestock and Poultry Slaughtering and Processing-Pigs (GB/T 19479-2004) during the slaughtering process. Immediately after euthanasia, SAT samples from the dorsal region of the last rib and VAT samples adjacent to the kidneys were collected using aseptic techniques and subsequently snap-frozen in liquid nitrogen for preservation. Eight-week-old male C57BL/6 mice were obtained from DOSSY Experimental Animals Co., Ltd. (Chengdu, China). All mice were maintained in an environment with controlled temperature and lighting, following a 12-h light/12-h dark schedule. To induce obesity via a high-fat diet, C57BL/6 mice were randomly allocated to receive either a standard diet or a 60% high-fat purified diet (XTHF60, Nanjing, China) for a duration of 12 weeks. After 2 weeks of high-fat adaptive feeding, Agomir-tRF-Gly or Agomir-NC were administered via intraperitoneal injection to the whole body and concurrently via in situ injection into iWAT and gWAT every 3 days, for a total of ten injections. This combined administration strategy was chosen on the basis of preliminary experiments showing that it produced the most robust and reproducible effects, while allowing simultaneous evaluation of systemic lipid metabolic regulation and depot-specific adipose responses within the same animals, thereby reducing inter-individual variation and the total number of animals used. All adipose tissues were rapidly stored in liquid nitrogen after collection for subsequent experiments. All experiment animals (pigs and mice) were approved by the Animal Protection and Ethics Committee of Sichuan Agricultural University (Chengdu, China, no. 20240461). Cell culture and induction of differentiation to adipocytes 3T3-L1 preadipocytes (cat. no. GNM25) and 293 T (cat. no. GNHu17) cells were obtained from the National Infrastructure of Cell Line Resources (NICR, Beijing, China). Proliferative cell cultures of 3T3-L1 preadipocytes and 293 T cells were grown in Dulbecco’s modified Eagle medium (DMEM; Gibco, USA) enriched with 10% fetal bovine serum and 1% penicillin–streptomycin (Gibco, USA). When 3T3-L1 preadipocytes reached 100% fusion, the culture was continued for 48 h in order to inhibit proliferation and initiate the differentiation process. At the end of contact inhibition, the proliferation phase medium was replaced with the induced differentiation medium containing insulin (10 μg/mL), 3-isobutyl-1-methylxanthine (0.5 mM), dexamethasone (1 μM), and rosiglitazone (1 μM). After 2 days of induced differentiation, the induced differentiation medium was replaced with the maintenance medium containing only insulin (10 μg/mL) and rosiglitazone (1 μM). The maintenance medium was changed every 2 days, and mature 3T3-L1 adipocytes were successfully obtained 8–10 days after the start of induction. All cells were cultured at 37 ℃ with 5% CO2. tRF and tiRNA Sequencing Adipose tissue RNA was isolated using RNAiso Plus (9109, Takara, Japan), then processed with an RNA Pretreatment Kit (Arraystar, USA) to eliminate modifications such as 3′-aminoacyl, 2′,3′-cyclic phosphate, 5′-OH, m1A, and m3C. The RNA samples were subsequently ligated with both 3′ and 5′ adapters, and cDNA synthesis was performed using Illumina’s RT and amplification reagents. PCR products ranging from 134 to 160 bp were excised from a PAGE gel and purified. The resulting libraries were quantified using an Agilent 2100 Bioanalyzer (Agilent, USA), diluted appropriately, and sequenced on an Illumina NextSeq 500 system with a 50 bp single-read configuration. Raw tRF and tiRNA sequencing data are available in the National Genomics Data Center (accession PRJNA899798). LC/MS nontargeted metabolomics Metabolite extraction was performed on 3T3-L1 adipocytes and porcine visceral adipose tissue using 1 mL of precooled methanol:acetonitrile:water (2:2:1) while subjecting samples to sonication for 1 h in an ice bath. After sonication, the samples were incubated at −20 ℃ for an additional hour, then centrifuged at 16,000 g at 4 ℃ for 20 min, and finally transferred to sampling vials for liquid chromatography–mass spectrometry (LC–MS) analysis. The LC–MS setup utilized a Shimadzu Nexera X2 LC-30AD system fitted with an Acquity UPLC HSS T3 column (1.8 μm, 2.1 mm × 50 mm, Waters) paired with a 5500 QTRAP triple quadrupole mass spectrometer (AB SCIEX). Metabolite detection was carried out in both electrospray negative- and positive-ionization modes, with quality control samples injected periodically throughout the run. Raw MRM data from MT1000 KIT metabolites were processed using MultiQuant 3.0.2 to obtain the peak area for each metabolite. Cell transfection and chemical drugs therapy The exogenous tRF-Gly mimics and inhibitor (lock nucleic acid modification, LNA) were synthesized at Genepharma (Shanghai, China). 3T3-L1 preadipocytes and 293 T cells were inoculated in 12-well plates at a density of 2 × 105 cells/mL and cultured for 24 h. The exogenous tRF-Gly mimics and inhibitor were transfected into cells at a final concentration of 20 nM (40806es, YEASEN). pcDNA3.1-RAC1 (OE-RAC1) and pcDNA3.1-RAC1 control (OE-Vector) were purchased from Tsingke Biotechnology Co., Ltd. (Chengdu, China). Transfection of RAC1 was performed using the Lipofectamine 3000 (Invitrogen, USA) liposome transfection method as described in the manufacturer’s instructions. The sequences of the relevant nucleotides are shown in Supplementary Tables S1 and S2. RNA extraction and quantitative real-time PCR (qRT-PCR) Following the manufacturer’s guidelines, total RNA was isolated from both adipose tissue and cell samples using RNAiso Plus (Takara, Japan). The RNA concentration for each sample was then determined using a NanoDrop 2000 (Thermo Fisher Scientific, USA). For reverse transcription, mRNA was converted to cDNA using the PrimeScript™ RT reagent kit (Takara, Japan), while tsRNA was processed with the Mir-X™ miRNA First-Strand Synthesis Kit (Takara, Japan). The cDNA synthesis protocol for mRNA included incubation at 42 ℃ for 2 min, 37 ℃ for 15 min, followed by a 5-s step at 85 ℃. For tsRNA, the reaction was carried out at 37 ℃ for 1 h and then heated at 85 ℃ for 5 min. The generated cDNA was subsequently subjected to qRT-PCR using TB Green PreMix EX Taq™ II (Takara, Japan) on a Bio-Rad CFX96 Real-Time PCR Detection System (Bio-Rad, USA). GAPDH and β-actin were employed as internal controls for mRNA, while U6 was used for tsRNA. Relative expression levels were calculated using the 2−∆∆Ct method [28], with all gene expressions normalized accordingly. Primer sequences used in the qRT-PCR experiments are provided in Supplementary Table S3. Cell proliferation assay To assess the proliferative capacity of 3T3-L1 preadipocytes, evaluations were conducted 24 h post-transfection with exogenous mimics. Cell proliferation was measured using the CCK-8 assay (Beyotime, Shanghai, China) and EdU incorporation assay (Ribobio, Guangzhou, China). Transfected cells were plated in 96-well plates, treated with 10 μL of CCK-8 solution, and incubated for 1.5 h before measuring absorbance at 450 nm with a Varioskan LUX microplate reader (Thermo Fisher, USA). This process was repeated at 0, 24, 48, and 72 h to track proliferation over time. For the EdU assay, cells in 96-well plates were exposed to 50 μM EdU for 2 h, following the manufacturer’s instructions. Fluorescence imaging was then performed to visualize proliferating cells. TdT-mediated dUTP nick-end labeling (TUNEL) staining 3T3-L1 preadipocytes transfected with mimics tRF-Gly, inhibitor tRF-Gly, or their corresponding negative controls were fixed with 4% paraformaldehyde (Beyotime, China) in phosphate-buffered saline (PBS) for 30 min at room temperature (RT). After fixation, cells were permeabilized with 0.3% Triton X-100 for 20 min and incubated with TUNEL reagent (Elabscience, China; cat. no. E-CK-A320) following the manufacturer’s instructions to label DNA strand breaks. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Beyotime, China). Fluorescence images were obtained using an OLYMPUS FV4000 confocal laser scanning microscope (Olympus, Tokyo, Japan), and the percentage of TUNEL-positive nuclei relative to total nuclei was quantified from randomly selected fields. Flow cytometry analysis of apoptosis Apoptosis of 3T3-L1 preadipocytes transfected with tRF-Gly mimics, tRF-Gly inhibitor, or their corresponding negative controls was assessed using an Annexin V-YSFluor™ 488/PI Apoptosis Detection Kit (40305ES, YEASEN) according to the manufacturer’s instructions. Briefly, the cells were collected by centrifugation at 1000 rpm for 5 min, washed twice with cold phosphate-buffered saline (PBS), and resuspended in 500 μL of binding buffer containing 5 μL Annexin V-YSFluor™ 488 and 10 μL PI (20 μg/mL). After incubation for 10 min in the dark at room temperature, the samples were analyzed on a Beckman flow cytometer (Beckman Coulter, USA) to quantify the percentage of apoptotic cells. Dual luciferase reporter assay The dual luciferase reporter assay was conducted to assess whether tRF-Gly mimics directly bind to the 3′ UTR of RAC1. For this purpose, the RAC1 3′ UTR was inserted into the pSiCheck2 vector by Tsingke Biotechnology Co., Ltd. (Chengdu, China). Subsequently, cells were seeded into 12-well plates at a density of 200,000 cells/mL and co-transfected with either control mimics (NC) or tRF-Gly mimics, along with either wild-type or mutated RAC1 plasmids, using Lipofectamine 3000 (Thermo Fisher, USA) for 48 h. The mimics were applied at a final concentration of 30 nM, and the plasmids were introduced at 100 ng per well. Luciferase activity was then measured following Promega’s protocol, with Renilla luciferase signals normalized to those of Photinus pyralis luciferase using a Varioskan LUX Multifunctional Enzyme Labeler (Thermo Fisher, USA). Western blot analysis Cells were lysed for 30 min using a RIPA solution (Beyotime, China) containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and 1 mM PMSF (Beyotime, China) to extract total cellular proteins. The protein concentration in each sample was then quantified with a BCA Protein Assay Kit (Beyotime, China). Following separation by SDS-PAGE, the proteins were transferred onto PVDF membranes. The membranes were incubated overnight at 4 ℃ with the primary antibodies, and afterward, they were treated with anti-rabbit or anti-mouse IgG secondary antibody for 1 h at room temperature. Finally, all PVDF membranes were incubated with enhanced chemiluminescence (ECL) reagent (SB-WB004, ShareBio) and subsequently visualized and imaged using the eBLOT Western Blot imaging system (Beijing, China). The corresponding antibodies were obtained from the following designated locations: anti-GAPDH (no. 14C10, 1:1000), anti-α-tubulin (no. 11H10, 1:1000), anti-β-tubulin (no. 2146, 1:1000), anti-RAC1 (no. 4651, 1:1000) and Argonaute 3 (no. 5054, 1:1000) were obtained from Cell Signaling Technology (Boston, USA). Anti-CEBPα (ab40764, 1:1000) and anti-PPARγ (ab272718, 1:1000) were obtained from Abcam (Cambridge, UK). Anti-GSK3β (no. A11731, 1:1000), anti-AXIN1 (no. A16019, 1:1000), anti-AXIN2 (no. A17021, 1:1000), anti-CK1α (no. A9308, 1:1000), and anti-JNK2 (no. A1251, 1:1000) were obtained from ABclonal (Wuhan, China). Anti-β-catenin (no. R23616, 1:1000) and anti-histone H3 (no. 250,011, 1:1000) were obtained from ZENBIO (Chengdu, China). Anti-P-β-catenin (Ser33/37) (no. HA721580, 1:1000) was obtained from HUABIO (Hangzhou, China). Protein blots were visualized by Touch Imager (eBlot, Shanghai, China). All protein band images were processed and analyzed using ImageJ software (1.52a version, USA). Northern blotting Northern blotting was used to verify the overexpression and knockdown effects of tRF-Gly. Initially, 4 μg of total RNA was heat-denatured at 95 ℃ for 2 min and then separated on a 12.5% urea acrylamide gel. The RNA was subsequently transferred to a pre-wetted Hybond-N + nylon membrane (Millipore). After the transfer, the membrane was subjected to UV crosslinking at 120 mJ using a Stratagene UV Stratalinker 1800 (SCIENTZ03-II, Ningbo, China) and subsequently hybridized with biotin-labeled DNA probes provided by Sangon Biotech (Shanghai, China). The sequences for these probes are detailed in Supplementary Table S4. RNA immunoprecipitation (RIP) assay Cells were rinsed with pre-chilled PBS, detached by scraping, and centrifuged at 2000 rpm for 5 min. Subsequently, cells were lysed in 1 mL of lysis buffer supplemented with RNase OUT (100 U/ml; Takara) and a protease inhibitor cocktail (PIC; Roche) on ice for 45 min. The lysate was then centrifuged at 12,000 g for 15 min at 4 ℃, and the supernatant was collected for protein quantification using the BCA assay. Samples were incubated with Antibody AGO3 (no. 5054 T, CST, Boston, USA) or control IgG (no. 2729, CST, Boston, USA) overnight at 4 ℃, then samples were incubated with 30 μL of protein G magnetic beads (Invitrogen) for 1 h at 4 ℃. Beads were washed three times in lysis buffer. RNA was extracted from beads using the Trizol method for qRT-PCR, and a portion was used for IP-WB to detect AGO3. The results of retrieved RNAs are presented as a percentage of the input. Bodipy staining and Oil Red O staining Cells were fixed in 4% paraformaldehyde (Beyotime, China) at room temperature for 1 h. Following fixation, the cells were rinsed 2–3 times with PBS and then stained with Bodipy (Thermo Fisher, USA) for 30 min. After an additional 2–3 washes, the nuclei were counterstained with DAPI (Beyotime, China) for 20 min at room temperature. Finally, the cells were examined using a fluorescence microscope, and the images were analyzed with ImageJ software (version 1.52a, USA). Fully differentiated mature adipocytes were fixed in 4% paraformaldehyde for 1 h, then Oil Red O staining was performed to evaluate intracellular lipid droplet accumulation. Following fixation, cells were gently washed 2–3 times with PBS and then rinsed with 60% isopropyl alcohol for 15 s to enhance neutral lipid staining. Subsequently, the cells were immersed in the Oil Red O staining solution for 15 min, in a light-protected environment, and later observed and photographed under a bright fluorescent microscope. Five random points were selected for each treatment group, and the lipid droplet area was quantified using ImageJ software (1.52a version, USA). Triglyceride (TG) and total cholesterol determination (T-CHO) The cell suspension was spun at 1000 rpm for 10 min, after which the supernatant was carefully discarded and the pellet washed 2–3 times with PBS. Following the washes, the cell precipitates were sonicated under ice water bath conditions to facilitate cell lysis. Subsequent assays were performed according to the protocol provided by Nanjing Jiancheng Bioengineering Institute. TG and T-CHO assays were respectively conducted at wavelengths of 500 nm and 510 nm. The absorbance of each well was measured using the Varioskan LUX Multifunctional Enzyme Labeler (Thermo Fisher, USA), and the triglyceride and total cholesterol content were calculated on the basis of predefined formulas. Bioinformatics analysis Visualization of raw tRF and tiRNA sequencing data generated by the NextSeq 500 Sequencing System using the Bioinformatics Visualization website (https://www.bioinformatics.com.cn/). In the R environment, online resources such as TargetScan (https://www.targetscan.org/) and miRDB (http://www.mirdb.org/) were used for the purpose of predicting target genes of tRF-Gly. Functional enrichment of target genes by MetaScape and analysis of signaling pathways involved in target genes using KEGG pathway. RNAfold WebServer (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi) was used to predict the cleavage site of tRF-Gly. Statistical analysis Statistical comparisons were performed using an unpaired two-tailed Student’s t-test for two-group analyses. All data were expressed as mean ± SEM and analyzed using GraphPad Prism 8.0.2 (San Diego, CA, USA). Unless indicated otherwise, experiments were performed in triplicate (n = 3), and significance was set at P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, or not significant [ns]). Results tRF-Gly is strongly associated with lipogenesis, including in pigs and mice Given that Sus scrofa is a valuable model organism for studying obesity pathogenesis [29], we used visceral adipose tissue (VAT) from representative obese (Qing Yu pig, QYP) and lean (Yorkshire pig, YP) pig breeds as experimental materials for preliminary tsRNA screening [14, 27]. H&E staining revealed significant differences in visceral adipocyte size between the two pig breeds (Fig. 1A). To investigate the potential differences in fatty acid metabolism of visceral fat between YP and QYP pigs, LC–MS nontargeted metabolomics was used to investigate the adipose lipid composition and distribution in the VAT of these two pigs. Principal component analysis (PCA) and orthogonal partial least squares–discriminant Analysis (OPLS-DA) highlighted distinct variations in adipose lipid composition between YP and QYP pigs (Supplementary Fig. S1A, B). Permutation testing confirmed the high reliability of the LC–MS nontargeted metabolomics results (Supplementary Fig. S1C). In the visceral fat samples from the two pig breeds, 83 metabolites that satisfied the criteria (VIP > 1, P < 0.05) exhibited significant differences. Heatmap cluster analysis of differential metabolites revealed distinct abundance patterns and type distributions of fatty acid composition between obese and lean pigs (Fig. 1B). In addition, the differential metabolites were predominantly concentrated in sphingolipids and phosphatidylacids (Fig. 1C). KEGG analysis revealed that these metabolites were mainly associated with metabolic processes, particularly glycerolipid metabolism (Fig. 1D). These results indicated that lipid metabolic pathways differed between QYP and YP and were closely associated with differences in body fat percentage and fat deposition. tRF-Gly expression is closely linked to lipogenesis. A VAT tissue sections from YPs and QYPs were stained using hematoxylin and eosin (H&E) (200×, 50 μm). The histogram presents the quantitative analysis of visceral adipocyte sizes in the two pig breeds. B The heatmap displays the differences in fatty acid expression of VAT between YPs and QYPs (n = 6). C The pie chart illustrates the differential metabolite types between the VAT of YPs and QYPs (n = 6). D KEGG enrichment analysis of significantly different metabolites (n = 6). E The volcano plot depicts differential tsRNA expression in VAT between YPs and QYPs, with an arrow highlighting the tRF-Gly position. F The bidirectional bar chart shows the top ten tsRNAs with the highest counts per million mapped reads (CPM) values in YPs and QYPs. G The expression of tRF-Gly in the VAT and SAT of YPs and QYPs was measured by qRT-PCR. H The cleavage site map of tRF-Gly derived from the secondary structure of tRNA-Gly. I The Venn diagram illustrates the conservation of tRF-Gly between pig and mouse species. J–L qRT-PCR of the relative expression level of tRF-Gly in mice fed either a low-fat diet (LFD) or a high-fat diet (HFD) (J), mice at different developmental stages (K), and at various differentiation time points of 3T3-L1 preadipocytes (L). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM Furthermore, to identify obesity-related tsRNAs, we analyzed the tRF and tiRNA sequencing data from previous studies and identified 151 significantly differentially expressed tsRNAs (|log2FC|≥ 2, P < 0.05). The volcano plot showed that 81 of these tsRNAs were highly expressed in QY pigs (Fig. 1E). In addition, the top ten tsRNAs in VAT were selected on the basis of CPM values, with tRF-Gly showing the highest expression abundance in the VAT of QY pigs (Fig. 1F). qRT-PCR was then performed to validate the relative expression of tRF-Gly in the VAT and SAT of lean and obese pigs (Fig. 1G). The RNAfold WebServer was used to predict the cleavage site of tRF-Gly, revealing that it originates from the 5′ end of tRF-Gly-GCC-1-1 and belongs to the tRF-5c type (Fig. 1H). Comparison of tRF and tiRNA sequencing data from mouse and pig adipose tissue revealed that tRF-Gly is a highly conserved tRF [14] (Fig. 1I). This finding provided a foundation for subsequent functional verification of tRF-Gly in both mouse-derived 3T3-L1 preadipocytes and mice. Furthermore, we analyzed the expression of tRF-Gly in mice fed either a low-fat diet (LFD) or a high-fat diet (HFD) (Fig. 1J), in mice at different developmental stages (21, 42, and 56 days; Fig. 1K), and in 3T3-L1 preadipocytes at different differentiation time points (2, 4, 6, and 8 days; Fig. 1L). The results showed a strong correlation between tRF-Gly expression and fat deposition, suggesting its potential involvement in the adipogenesis process both in vivo and in vitro. tRF-Gly modulates proliferation and apoptosis in 3T3-L1 preadipocytes Fat deposition is highly correlated with the multiplication of preadipocytes. Enhanced preadipocyte proliferation lays the cellular groundwork for later differentiation. To investigate how tRF-Gly influences lipogenesis, we initially analyzed the Gene Ontology (GO) terms associated with its predicted potential targets. Our findings indicated that the predicted targets of tRF-Gly are pivotal for modulating cell proliferation, apoptosis, differentiation, and lipid metabolism (Fig. 2A). For functional verification, we synthesized and transfected exogenous mimics and inhibitors of tRF-Gly into 3T3-L1 preadipocytes. Northern blot analysis was employed to assess tRF-Gly transfection efficiency, with qRT-PCR providing additional validation (Fig. 2B, C). Sequence alignment of the Sanger sequencing results verified that the PCR product corresponded to tRF-Gly (Fig. 2D). We further analyzed the mRNA levels of genes related to proliferation and apoptosis following the successful overexpression or knockdown of tRF-Gly in 3T3-L1 preadipocytes. The findings showed that tRF-Gly overexpression upregulated the mRNA levels of cell-cycle-related genes PcnA and Ccnd1, along with the anti-apoptotic gene Bcl2, while downregulating the pro-apoptotic gene Bax. Conversely, tRF-Gly knockout led to a reduction in mRNA expression levels of cell cycle genes Pcna and Ccnd1, as well as the anti-apoptotic gene Bcl2, with no significant effect on the mRNA expression level of the pro-apoptotic gene Bax (Fig. 2E). The CCK-8 cell viability assay and EdU assay further verified that overexpression of tRF-Gly accelerated cell proliferation, while knockout of tRF-Gly inhibited cell proliferation (Fig. 2F, G). Consistent with gene expression analysis, TUNEL staining and annexin V/PI flow cytometry further confirmed that tRF-Gly reduced apoptosis in 3T3-L1 preadipocytes, while its inhibition enhanced apoptosis (Fig. 2H, I). Together, these findings suggest that tRF-Gly plays a regulatory role in the proliferation and apoptosis of 3T3-L1 preadipocytes. tRF-Gly regulates 3T3-L1 preadipocytes proliferation and apoptosis. A GO enrichment analysis of tRF-Gly predicted target genes. B Evaluation of tRF-Gly transfection efficiency in proliferative phase 3T3-L1 preadipocytes using northern blot. Quantification of tRF-Gly and tRNA-Gly-GCC expression in 3T3-L1 preadipocytes by northern blot analysis, with U6 as the internal control. C The expression level of tRF-Gly in the proliferative phase of 3T3-L1 preadipocytes after transfection of the tRF-Gly mimic and inhibitor. D The Sanger sequencing of the PCR products of tRF-Gly. E The mRNA levels of proliferation and apoptosis-related genes in 3T3-L1 preadipocytes were analyzed by qRT-PCR after overexpression or knockdown of tRF-Gly. F The CCK-8 cell viability assay for the effect of tRF-Gly on the proliferation of 3T3-L1 preadipocytes. The detection time was 0, 24, 48, and 72 h after transfection. G The EdU assay for the effect of tRF-Gly overexpression or knockdown on the proliferation of 3T3-L1 preadipocytes (100×, 100 μm). H TUNEL staining showing apoptotic cells after transfection with tRF-Gly mimic or inhibitor (200× , 50 μm). I Annexin V/PI flow cytometry assay of apoptosis in 3T3-L1 preadipocytes after tRF-Gly overexpression or inhibition, with the corresponding quantification of apoptosis index. Bar graphs illustrating the percentage of viable (Q4), early apoptotic (Q3), late apoptotic (Q2), and necrotic (Q1) cells according to annexin V FITC/PI staining. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM tRF-Gly regulates adipogenic differentiation in 3T3-L1 adipocytes Considering that the predicted potential target genes of tRF-Gly are also enriched in differentiation and lipid metabolism, we further examined how altering tRF-Gly levels through overexpression and knockdown impacts the differentiation of 3T3-L1 adipocytes. Similarly, we assessed tRF-Gly delivery in differentiated 3T3-L1 adipocytes using northern blotting, with qRT-PCR providing additional confirmation. (Fig. 3A, B). After tRF-Gly was successfully modulated through overexpression or silencing, we measured Cebpα and Pparγ, which are pivotal markers of adipogenic differentiation in 3T3-L1 adipocytes, using qRT-PCR. Overexpression of tRF-Gly resulted in increased levels of Cebpα and Pparγ, while its knockdown led to a decrease in these markers (Fig. 3C). Protein analysis via western blot confirmed that tRF-Gly modulated CEBPα and PPARγ levels, with the observed changes mirroring those at the transcriptional level (Fig. 3D, E). We further evaluated lipid droplet accumulation during 3T3-L1 adipocyte differentiation following tRF-Gly modulation, employing Bodipy and Oil Red O staining. The results from both staining were consistent. Concretely, overexpression of tRF-Gly promoted lipid droplet accumulation during differentiation, whereas interference with tRF-Gly inhibited lipid droplet formation (Fig. 3F, G). Biochemical analysis indicated that upregulating tRF-Gly enhanced triglyceride and total cholesterol secretion during 3T3-L1 cell differentiation, whereas its inhibition yielded the opposite effect (Fig. 3H, I). We also performed LC–MS nontargeted metabolomics on 3T3-L1 differentiated adipocytes overexpressing tRF-Gly. The results showed that 32 significant metabolites were identified after overexpression of tRF-Gly in 3T3-L1 differentiated adipocytes. tRF-Gly significantly altered the distribution of lipid metabolites, reflecting a metabolic shift from energy demand to lipid storage during cell differentiation (Fig. 3J). In summary, tRF-Gly modulated the maturation process of 3T3-L1 adipocytes. tRF-Gly promotes lipid differentiation during the differentiation of 3T3-L1 adipocytes. A Evaluation of tRF-Gly transfection efficiency in 3T3-L1 differentiated adipocytes using northern blot. Quantification of tRF-Gly and tRNA-Gly-GCC level in 3T3-L1 adipocytes using U6 as the internal control. B tRF-Gly expression during 3T3-L1 adipocyte differentiation was analyzed following transfection with its mimic and inhibitor. C Cebpα and Pparγ mRNA expression was determined by qRT-PCR following tRF-Gly modulation. D The protein levels of CEBPα and PPARγ in 3T3-L1 adipocytes with overexpression of tRF-Gly were assessed by western blotting. The histogram quantifies the change in target protein content. E The protein levels of CEBPα and PPARγ in 3T3-L1 adipocytes with knockdown of tRF-Gly were assessed by western blotting. F The accumulation of lipid droplets in 3T3-L1 adipocytes following overexpression or knockdown of tRF-Gly was assessed by Bodipy staining. The histogram on the right quantifies the fluorescence intensity of Bodipy-stained lipid droplets (200×, 50 μm). G The accumulation of lipid droplets in 3T3-L1 adipocytes following overexpression or knockdown of tRF-Gly was assessed by Oil Red O staining. The histogram on the right quantifies the accumulation of lipid droplets stained with Oil Red O. H, I The changes in triglyceride (H) and total cholesterol (I) accumulation levels in 3T3-L1 adipocytes following overexpression or knockdown of tRF-Gly were measured using biochemical assays. J The differential changes in significant metabolites in 3T3-L1 adipocyte overexpressing tRF-Gly were analyzed by LC–MS nontargeted metabolomics. The numbers shown next to each bar indicate the corresponding Log2 (fold change) values of metabolites (mimics tRF-Gly versus mimics NC). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM tRF-Gly targets Rac1 to inhibit the Wnt/β-catenin signaling pathway Bioinformatic prediction of tRF-Gly targets was performed using TargetScan, miRDB, and tRFtarget 2.0, with DAVID-based functional analysis revealing potential pathways involved in lipid deposition regulation. The Wnt/β-catenin pathway emerged as the central regulatory network for tRF-Gly targets through comprehensive KEGG pathway annotation (Fig. 4A). The chord diagram depicted the ten predicted target genes of tRF-Gly associated with the Wnt/β-catenin signaling pathway (Fig. 4B). Numerous studies have shown that tsRNAs can exert their functions via a silencing mechanism analogous to that of microRNAs [11, 30]. We used qRT-PCR to validate ten predicted target genes linked to the Wnt/β-catenin signaling pathway (note that the Cq values of Nkd2, Ctnnd2, Camk2d, and Lgr6 were > 35, indicating no expression, and thus these genes were excluded from the graph). The results revealed that only Rac1 exhibited a consistent increase or decrease in expression following overexpression or knockdown of tRF-Gly (Fig. 4C; Supplementary Fig. S2A). Western blot analysis revealed that the protein levels of RAC1 were also regulated by tRF-Gly, with changes consistent with the transcriptional expression levels (Fig. 4D). Further, 293 T cells were co-transfected with either tRF-Gly mimics or a negative control, together with the psiCHECK2 plasmid containing either the wild-type Rac1 3′ UTR or a mutated version of the Rac1 3′ UTR. After 36 h, luciferase activity was assessed, and the results indicated that tRF-Gly significantly reduced luciferase activity in cells transfected with the psiCHECK2-Rac1 vector, compared with those transfected with the mutated construct (Fig. 4E, F). In addition, we proposed that tRF-Gly might regulate gene expression via Argonaute (AGO) family proteins, similar to the mechanism employed by microRNAs [11, 30–32]. To determine which AGO protein is involved in the tRF-Gly-mediated regulation of the target gene Rac1, we knocked down the expression of AGO1–4 by synthesizing their corresponding siRNAs (Supplementary Fig. S2B). Subsequently, we assessed the changes in luminescence intensity using the dual luciferase assay. The results showed that AGO3 interference reversed the reduction in dual luciferase activity caused by tRF-Gly binding to psiCHECK2-Rac1 (Fig. 4G). To further validate AGO3’s participation in Rac1 regulation by tRF-Gly, RIP-qPCR analysis demonstrated a significant enrichment of AGO3 within both tRF-Gly and Rac1 complexes (Fig. 4H). Molecular docking suggested that tRF-Gly binds AGO3 via the region GCAUUGG (nucleotides 1–7), which is predicted to form a stem–loop structure (Fig. 4I). These results indicated that tRF-Gly regulates Rac1 expression by binding to AGO3. tRF-Gly is involved in the Wnt/β-catenin signaling pathway by regulating RAC1/JNK2/β-catenin transport complex expression. A KEGG enrichment analysis of tRF-Gly predicted target genes. The numbers in the bars indicate the counts of predicted target genes belonging to each pathway (e.g., 18 genes were annotated to the Wnt signaling pathway). B The chord diagram shows the ten predicted target genes of tRF-Gly associated with the Wnt/β-catenin signaling pathway. C The mRNA expression level of Rac1 in the differentiation stage of 3T3-L1 adipocytes after transfection with the tRF-Gly mimic and inhibitor. D The protein levels of RAC1 in 3T3-L1 adipocytes with overexpression or knockdown of tRF-Gly were assessed by western blotting. The histogram on the right quantifies the change in RAC1 protein content. E The luciferase activity of psiCHECK2-Rac1 was significantly decreased by the 5′ tRF-Gly mimics in 293T cells. F Schematic representation illustrating the predicted binding site of 5′ tRF-Gly with the seed region in the Rac1 3′ UTR, along with the corresponding mutation region. G The dual luciferase assay showed that interference with AGO3 expression restored the decrease in dual luciferase activity caused by the binding of tRF-Gly to psiCHECK2-Rac1. H The relationship between tRF-Gly, Rac1, and AGO3 was assessed on the basis of RIP-qPCR analysis. I Schematic diagram depicting molecular docking between tRF-Gly and AGO3. J The activation of the Wnt/β-catenin signaling pathway in 3T3-L1 adipocytes with overexpression or knockdown of tRF-Gly was assessed by western blotting. The histogram on the right quantifies the changes in target protein levels. K The nuclear translocation of β-catenin in 3T3-L1 adipocytes with overexpression or knockdown of tRF-Gly was assessed by western blotting. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM We further explored if modulating tRF-Gly expression in 3T3-L1 adipocytes influenced the activity of the Wnt/β-catenin signaling cascade. We explored the effect of tRF-Gly on β-catenin, the key downstream effector protein of the Wnt/β-catenin signaling pathway, using western blot analysis. The results showed that overexpression of tRF-Gly led to a decrease in total β-catenin protein levels in 3T3-L1 adipocytes, whereas interference with tRF-Gly caused an increase in total β-catenin protein levels. For the Wnt/β-catenin signaling cascade to be activated, β-catenin must move from the cytoplasm into the nucleus. In contrast, when β-catenin is phosphorylated, it is swiftly degraded by complexes such as GSK3β, CK1α, and AXIN1, which prevents its entry into the nucleus and consequently suppresses Wnt signaling activation. Western blot results also revealed that overexpression of tRF-Gly increased the levels of phosphorylated β-catenin (P-β-catenin) in 3T3-L1 adipocytes, along with an upregulation of the degradation complex proteins. In contrast, tRF-Gly interference reduced the levels of P-β-catenin and the content of degradation complex proteins (Fig. 4J). However, qRT-PCR results showed that overexpression or knockdown of tRF-Gly had no significant effect on the transcription levels of β-catenin and its degradation complex components (Supplementary Fig. S2C). Earlier studies have indicated that the RAC1/JNK2/β-catenin transport complex is essential for β-catenin’s nuclear translocation [33–37]. We also found that tRF-Gly modulated the protein expression of components in the RAC1/JNK2/β-catenin transport complex (Fig. 4J). Further, we separated the cytoplasmic and nuclear components of 3T3-L1 adipocytes and performed immunoblotting for β-catenin. The results showed that the nuclear translocation of β-catenin was significantly inhibited upon overexpression of tRF-Gly, while no significant promotion of nuclear translocation of β-catenin was observed upon interference with tRF-Gly (Fig. 4K). Collectively, our findings indicate that tRF-Gly influences the intracellular distribution of β-catenin by modulating the RAC1/JNK2/β-catenin transport complex, ultimately affecting the Wnt/β-catenin signaling pathway. Furthermore, we conducted rescue experiments to determine whether tRF-Gly could counteract the inhibition of adipogenic differentiation induced by the Wnt/β-catenin signaling pathway agonist SKL2001 [38]. Western blot analysis showed that SKL2001 treatment resulted in a marked decrease in CEBPα and PPARγ levels, along with a significant increase in β-catenin expression, a critical component of the Wnt/β-catenin signaling pathway. Notably, tRF-Gly overexpression restored CEBPα and PPARγ protein levels in 3T3-L1 cells while concurrently suppressing β-catenin expression (Supplementary Fig. S2D). Consistently, Bodipy staining demonstrated that tRF-Gly overexpression effectively reversed the SKL2001-induced suppression of lipid droplet formation during 3T3-L1 preadipocyte differentiation (Supplementary Fig. S2E). These findings were further corroborated by triglyceride biochemical assays and phenotypic observations, reinforcing the role of tRF-Gly in modulating adipogenesis through Wnt/β-catenin signaling (Supplementary Fig. S2F). Taken together, these results demonstrate that tRF-Gly is essential for adipogenic differentiation through its inhibition of the Wnt/β-catenin signaling pathway. RAC1 regulates adipogenesis in 3T3-L1 adipocytes via Wnt/β-catenin signaling pathway The data showed that tRF-Gly regulated the Wnt signaling pathway through RAC1, which in turn affected adipogenic differentiation in 3T3-L1 adipocytes. To investigate RAC1’s role in modulating the Wnt signaling pathway in 3T3-L1 adipocytes, we generated a RAC1-pcDNA3.1 overexpression plasmid and si-RAC1. Functional assays were then conducted by either upregulating or silencing Rac1 in these cells. The results of qRT-PCR showed successful overexpression or knockdown of Rac1 at the transcriptional level (Fig. 5A). Western Blot analysis further confirmed that RAC1 protein levels were successfully overexpressed or knocked down in 3T3-L1 adipocytes (Fig. 5B). Subsequently, we assessed the changes in marker genes for adipogenic differentiation in 3T3-L1 adipocytes. qRT-PCR results revealed that overexpression of RAC1 significantly inhibited adipogenic differentiation, while knockdown of RAC1 expression promoted adipogenic differentiation in 3T3-L1 adipocytes (Fig. 5C). Furthermore, we examined the expression of RAC1/JNK2/β-catenin complex related transcripts. The results showed that changes in RAC1 expression in 3T3-L1 adipocytes influenced Jnk-cascade-related genes, as well as Wnt/β-catenin-signaling-pathway-related genes (Fig. 5D, E). Western blot analysis confirmed these findings, showing that RAC1 also affected the levels of key adipogenic differentiation proteins, including CEBPα and PPARγ, as well as JNK2 and Wnt signaling-pathway-related proteins (including AXIN1, β-catenin, GSK3β, and CK1α) in 3T3-L1 adipocytes (Fig. 5F, G). Bodipy staining was used to evaluate the impact of RAC1 overexpression or knockdown on lipid droplet accumulation in 3T3-L1 adipocytes during differentiation. The results showed that overexpression of RAC1 inhibited lipid droplet accumulation, while knockdown of RAC1 promoted lipid droplet accumulation (Fig. 5H). Biochemical analysis of TG and T-CHO levels further confirmed the role of RAC1 in regulating adipogenic differentiation in 3T3-L1 adipocytes (Fig. 5I, J). In summary, these findings indicate that RAC1 modulated adipogenic differentiation in 3T3-L1 adipocytes via the Wnt/β-catenin signaling pathway. RAC1 governs the differentiation of 3T3-L1 adipocytes via the Wnt/β-catenin pathway. A The efficiency of RAC1 overexpression or knockdown was evaluated at the mRNA level by qRT-PCR. B Western blot analysis was utilized to evaluate RAC1 levels following its overexpression or knockdown in 3T3-L1 adipocytes. C The changes in adipogenic differentiation marker genes were assessed by qRT-PCR. D The changes in JNK cascade reaction marker genes were assessed by qRT-PCR. E Gene expression related to the Wnt/β-catenin pathway was measured in differentiating 3T3-L1 adipocytes following RAC1 overexpression or knockdown. F Protein expression associated with adipogenesis, JNK2, and the Wnt/β-catenin pathway was evaluated in differentiating 3T3-L1 adipocytes following RAC1 overexpression or knockdown. G The histogram shows the quantification of the immunoblotting bands from F. H Lipid droplet accumulation in differentiating 3T3-L1 adipocytes was evaluated via Bodipy staining following RAC1 overexpression or knockdown, with the accompanying histogram quantifying the fluorescence intensity. I, J The changes of triglyceride (I) and total cholesterol (J) accumulation levels in 3T3-L1 adipocytes following overexpression or knockdown of RAC1 were measured using biochemical assays. K Representative Bodipy staining of lipid droplet accumulation in 3T3-L1 adipocytes following tRF-Gly overexpression under RAC1 overexpression treatment (200×, 50 μm). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM tRF-Gly promotes obesity development in vivo Since adipogenic differentiation is a key step in fat deposition and obesity, we further explored the role of tRF-Gly in high-fat-diet-induced obesity in mice. The above data confirmed that tRF-Gly regulates the Wnt signaling pathway through the RAC1/Jnk2/β-catenin complex in vitro, thereby influencing adipogenic differentiation of 3T3-L1 adipocytes. To validate that tRF-Gly promotes fat deposition in vivo, we synthesized sulfo-Cyanine5 (CY5)-labeled Agomir-NC and Agomir-tRF-Gly, and injected them intraperitoneally into the whole body and in situ into the iWAT and gWAT every 3 days (Fig. 6A). Then, we detected the overexpression of tRF-Gly in iWAT, gWAT, and liver by qRT-PCR. Results demonstrated a significant increase in tRF-Gly expression in the iWAT, gWAT, and liver of mice administered Agomir-tRF-Gly (Fig. 6B). At the same time, the results of frozen section immunohistochemistry showed that tRF-Gly was successfully enriched in the iWAT (Fig. 6C). Four weeks after Agomir injection, mice in the Agomir-tRF-Gly group showed a marked increase in body weight relative to the other two groups (Fig. 6D). By the tenth week post-injection, mice in the Agomir-tRF-Gly group showed even more pronounced weight gain and obesity compared with the other groups (Fig. 6E). Similarly, statistical analysis of iWAT, gWAT, and BAT weights across the three groups showed that the Agomir-tRF-Gly group experienced a notably greater increase in these adipose tissues (Fig. 6F). The results of H&E staining showed that overexpression of tRF-Gly promoted the enlargement of adipocytes in both iWAT and gWAT, indicating that tRF-Gly positively regulated fat deposition in mice (Fig. 6G). Oil Red O staining was performed to assess lipid droplet accumulation in hepatocytes, and the results showed a significant increase in lipid droplet content in the liver of the Agomir group (Fig. 6H). Consistent with the phenotypic results, serum biochemical analysis showed that the levels of TG and T-CHO in the serum of mice in the Agomir-tRF-Gly group were significantly higher than those in the other groups (Fig. 6I, J). tRF-Gly intensified high-fat-diet-induced obesity in C57BL/6 mice. A Grouping of mice injected with Agomir. The high-fat-diet-induced obesity model was established using 8-week-old C57BL/6 J mice. After a 2-week adaptation period on a high-fat diet, mice received intraperitoneal and in situ injections of Agomir-NC or Agomir-tRF-Gly (5 nmol/mouse) into iWAT and gWAT every 3 days, for a total of ten injections. B After Agomir-tRF-Gly injection, the expression levels of tRF-Gly in iWAT, gWAT, and liver were measured by qRT-PCR (n = 6 mice). C Immunohistochemistry of frozen sections showed successful enrichment of tRF-Gly in iWAT. D The line chart of body weight changes after Agomir injection in HFD, Agomir-NC, or Agomir-tRF-Gly mice (n = 6 mice). E Comparison of body size among HFD, AgomirNC, and Agomir-tRF-Gly mice (n = 6 mice). F Statistical analysis of iWAT, gWAT, and BAT adipose tissue weights in HFD, Agomir-NC, and Agomir-tRF-Gly mice. G Representative H&E-stained sections of iWAT and gWAT tissues from mice in the HFD, Agomir-NC, and Agomir-tRF-Gly groups (200×, 50 μm). The histogram presents the quantitative analysis of adipocyte size in the iWAT and gWAT of the three groups of mice. H Oil Red O staining of liver tissue from HFD, Agomir-NC, and Agomir-tRF-Gly (200×, 50 μm). The histogram shows the quantitative analysis of lipid droplet accumulation in the liver, indicating a significant increase in the Agomir-tRF-Gly group. I, J The changes in triglyceride (I) and total cholesterol (J) levels in the serum of mice from the HFD, Agomir-NC, and Agomir-tRF-Gly groups were measured using biochemical assays. The results showed significantly higher levels of TG and T-CHO in the Agomir-tRF-Gly group compared with the other groups. K The expression levels of the target genes associated with the Wnt/β-catenin signaling pathway, including Cebpα, Pparγ, Rac1, Jnk2, β-catenin, Axin1, Ck1α, and Gsk3β, were measured in the iWAT of HFD, Agomir-NC, and Agomir-tRF-Gly groups. L Western blot analysis assessed Wnt/β-catenin pathway activation in iWAT from Agomir-tRF-Gly-treated mice by measuring the levels of adipogenesis-related proteins, the JNK2/β-catenin transport complex, and degradation complex components (AXIN1, CK1α, and GSK3β). M The histogram shows the quantification of the immunoblotting bands from L. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All results above are representative of three independent experiments and presented as the means ± SEM Furthermore, we investigated whether tRF-Gly regulates fat deposition in a manner consistent with the in vitro findings, specifically through the RAC1/JNK2/β-catenin transport complex to modulate the Wnt/β-catenin signaling pathway. In iWAT, the expression levels of fat deposition marker genes CEBPα and PPARγ were significantly increased in the Agomir-tRF-Gly group, indicating a more pronounced obese phenotype. In addition, in the Agomir-tRF-Gly group, the expression of the target gene RAC1 for tRF-Gly, as well as the transport complex JNK2 and β-catenin, were significantly reduced. Conversely, the transcripts of AXIN1, CK1α, and GSK3β, which are associated with the degradation complex of the Wnt/β-catenin signaling pathway, were significantly increased (Fig. 6K). At the protein level, we assessed the changes in adipogenesis-related proteins, transport complexes (JNK2/β-catenin), and degradation complexes (AXIN1, CK1α, and GSK3β) by western blotting, which were consistent with the in vitro results. In addition, the levels of phosphorylated β-catenin (Ser33/37) indicated that tRF-Gly could inhibit the activation of the Wnt/β-catenin signaling pathway by promoting the ubiquitination degradation of β-catenin in vivo, thereby facilitating the fat deposition process (Fig. 6L, M). In summary, these data strongly suggest that tRF-Gly regulates the Wnt/β-catenin signaling pathway through the RAC1/JNK2/β-catenin transport complex, thereby playing a key role in the regulation of fat deposition in mice. Discussion Currently, the precise function and regulatory pathways of tsRNAs in obesity and lipid accumulation are still poorly understood. Although previous research has suggested that tsRNAs may be involved in metabolic regulation, the exact molecular mechanisms through which they modulate adipogenesis are largely unexplored [12–14]. Our research findings have pinpointed tRF-Gly as a groundbreaking controller of fat formation, highlighting its crucial part in storing fat by fine-tuning the Wnt/β-catenin signaling route through the RAC1/JNK2/β-catenin trafficking machinery. In vitro and in vivo research confirmed the significant enhancement of adipocyte differentiation and fat deposition by tRF-Gly. This study not only uncovers the regulatory mechanism of tRF-Gly in adipogenesis but also highlights its potential as a novel target for obesity-related therapies. Previously thought to simply arise from tRNA breakdown, tsRNAs are now recognized as distinct small non-coding RNAs possessing significant functions. Emerging data indicate that tsRNAs, such as 5′tiRNA-His-GTG [11], tRF-3013b [31], and tRFGlnCTG [39] have been shown to critically modulate various biological processes, including cell differentiation, proliferation, and metabolic functions. Emerging evidence has implicated tsRNAs as epigenetic modulators of adipogenesis [12–14, 27], yet their mechanistic basis remains poorly defined. To investigate the epigenetic regulatory network involved in fat deposition, we screened potential tsRNAs from the tRF and tiRNA sequencing data of visceral fat in obese and lean pigs for subsequent functional verification. This screening identified tRF-Gly as the most differentially expressed tsRNA, exhibiting cross-species conservation between mouse and pig adipocytes, as evidenced by Clustal Omega sequence alignment. The evolutionary conservation of tRF-Gly motivated our functional investigation using the murine 3T3-L1 preadipocyte for adipogenesis research. Our findings identified tRF-Gly as a key regulator of fat formation, demonstrating its significant role in both in vivo and in vitro fat accumulation processes. The process of fat deposition is intricately linked to the proliferation of preadipocytes, as an increase in proliferating preadipocytes provides the cellular foundation for subsequent differentiation and fat accumulation. This research highlights the pivotal function of tRF-Gly in modulating the expansion of 3T3-L1 precursor fat cells. Our results show that overexpression of tRF-Gly accelerates cell proliferation, as evidenced by increased mRNA levels of cell cycle genes such as PCNA and cyclin D1, and enhanced cell viability in CCK-8 and EdU assays. Conversely, knockdown of tRF-Gly led to reduced proliferation and downregulated expression of these key cell cycle genes. The modulation of cell proliferation by tRF-Gly may provide a cellular foundation for enhanced adipogenesis, highlighting its potential as a key regulator of adipocyte differentiation. Future research should dissect tRF-Gly’s precise molecular actions on cell proliferation and its interplay with adipogenic signaling cascades. Moreover, fat deposition is tightly associated with preadipocyte differentiation, as the maturation of these cells into adipocytes is crucial for fat accumulation. Our studies on 3T3-L1 preadipocytes during differentiation revealed that tRF-Gly enhances fat cell development by regulating critical markers such as CEBPα and PPARγ, which play pivotal roles in lipid storage [40]. During adipocyte differentiation, the expression of triglyceride and fatty-acid-related synthetases is tightly regulated by key transcription factors, including CEBPα and PPARγ [41, 42]. Similarly, our study demonstrated that overexpression of tRF-Gly led to an upregulation of CEBPα and PPARγ at both the transcriptional and translational levels, promoting lipid droplet formation and enhancing triglyceride accumulation. Conversely, silencing tRF-Gly significantly decreased the expression of CEBPα and PPARγ, which in turn hampered the formation of lipid droplets and underscored its essential role in adipogenesis regulation. Furthermore, in vivo, the administration of Agomir-tRF-Gly led to significant fat accumulation in the iWAT and gWAT of mice. Although BAT thermogenesis and WAT browning are highly relevant to metabolic regulation, they were not the focus of the present study, which was primarily designed to explore fat deposition. Histological analysis showed that adipocytes in these tissues were significantly enlarged, indicating that tRF-Gly promotes adipocyte hypertrophy. In addition, serum biochemical assays revealed elevated levels of triglycerides (TG) and total cholesterol (T-CHO) in the Agomir-tRF-Gly group, further supporting the role of tRF-Gly in lipid accumulation. These findings collectively establish tRF-Gly as a pivotal regulator of fat cell development and imply its potential as a therapeutic focus for combating obesity and its associated metabolic disturbances. Mechanistically, we found that tRF-Gly regulates the expression of the target gene RAC1. Specifically, tsRNAs, like miRNAs, can modulate protein translation by interacting with translation initiation factors or affect transcription by binding to AGO proteins [31, 32, 43]. Previous studies have shown that tRF-5 type tRFs predominantly interact with AGO1, AGO3, and AGO4, with minimal interaction with AGO2 [44]. In our study, although tRF-Gly also belongs to the tRF-5 type, we still performed dual luciferase reporter gene assays using four AGO proteins (AGO1, AGO2, AGO3, and AGO4) in combination with tRF-Gly. The results indicated that the reduction in dual fluorescence intensity, owing to the specific interaction between tRF-Gly and its target gene RAC1, was effectively reversed only when AGO3 expression was knocked down. This suggests that tRF-Gly specifically interacts with AGO3, thereby facilitating the silencing of the target gene RAC1. It is worth noting that RAC1, a small GTPase belonging to the Rho family, has been linked to putting the brakes on PPARγ and CEBPα expression, as well as lipid droplet accumulation in 3T3-L1 cells [45, 46]. To further investigate how tRF-Gly mediates fat deposition through RAC1, we first conducted pathway enrichment analysis of its target genes. The findings indicated a robust link between tRF-Gly and the Wnt/β-catenin signaling axis. We demonstrated that overexpression or inhibition of tRF-Gly in vivo and in vitro will correspondingly alter the expression of genes and proteins tied to the Wnt/β-catenin signaling cascade. Previous studies have highlighted the role of the Wnt/β-catenin pathway in adipogenesis and obesity [16, 47]. For example, Wnt/β-catenin pathway activation inhibits adipocyte differentiation, whereas its blockade enhances fat storage. Moreover, β-catenin’s regulation of stability and breakdown via its engagement with the degradation machinery is pivotal in modulating adipocyte activity [22, 48]. In this study, we demonstrated that tRF-Gly hastens β-catenin breakdown, suppresses Wnt/β-catenin signals, and boosts lipogenesis. Mechanistically, we found that tRF-Gly blocks β-catenin nuclear transport via the RAC1/JNK2/β-catenin complex, modulating Wnt/β-catenin signaling. In addition, the upregulation of degradation complex components (AXIN1, CK1α, and GSK3β) likely enhances β-catenin breakdown, further modulating adipogenesis. Conclusions This research offers groundbreaking revelations about how tRF-Gly influences fat accumulation via the Wnt/β-catenin pathway, particularly emphasizing its pivotal role in fat storage through its modulatory effect on Wnt/β-catenin signaling via the RAC1/JNK2/β-catenin transport machinery. Collectively, the evidence indicates tRF-Gly may represent a viable target in addressing obesity and associated metabolic dysfunction. Future studies should explore the detailed molecular mechanisms by which tRF-Gly modulates Wnt signaling and its potential as a biomarker for obesity. Supplementary Information Supplementary Material 1. Acknowledgements We thank Linghui Wang (Bonn University) and Yanhao Qiu (University of Barcelona) for insightful discussions and generous support. Abbreviations CCAAT enhancer binding protein alpha Peroxisome proliferator activated receptor gamma White adipose tissue Visceral adipose tissue Subcutaneous adipose tissue Glycogen synthase kinase 3 beta Casein kinase 1 alpha 1 Qing yu pig Yorkshire pig Inguinal white adipose tissue Gonadal white adipose tissue Brown adipose tissue Mitogen-activated protein kinase 9 Rac family small GTPase 1 Author contributions Y.L. conceptualized the study, performed investigations and visualizations, provided resources, supervised the project, and drafted and revised the manuscript. M.G. and K.W. carried out investigations and visualizations, provided resources, and drafted and revised the manuscript. T.L., Y.Y., and X.Z. conducted investigations, provided resources, and drafted the manuscript. X.Z., D.C., X.W., and J.M. provided resources and reviewed and edited the manuscript. L.N., Y.Z., L.C., X.Z., and Y.W. performed formal analysis, curated data, and contributed to conceptualization. M.L., L.Z., and L.S. contributed to writing, review and editing, performed formal analysis and data curation, conceptualized the study, acquired funding, and supervised the project. All authors read and approved the final manuscript. Funding This work was supported by grants from National Natural Science Foundation of China (32573175, 32421005), the National Key Research and Development Program of China (2024YFF1000201), the Sichuan Science and Technology Program (2021YFYZ0007, 2021ZDZX0008), the Pig Industry Technology System Innovation Team of Sichuan Province (SCCXTD-2025–8), and the China Agriculture Research System (CARS-35). Data availability The raw sequencing data of tRF and tiRNA sequencing was deposited in the National Genomics Data Center accession PRJNA899798 and 10.3390/ani12243561. Any additional raw data that backs up the findings of this study is available from the corresponding author upon request. Declarations Ethics approval and consent to participate All procedures involving pigs and mice were approved on 2 July 2024 by the Animal Protection and Ethics Committee of Sichuan Agricultural University (Chengdu, China; approval no. 20240461) and were conducted in accordance with the International Council for Laboratory Animal Science (ICLAS) guidelines as well as the welfare and ethical standards of Sichuan Agricultural University’s Animal Management Committee. Consent for publication All authors have thoroughly reviewed this manuscript and kindly request its exclusive consideration for publication in Cellular & Molecular Biology Letters. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yuhang Lei, Mailin Gan and Kai Wang have contributed equally to this work. Contributor Information Li Zhu, Email: zhuli@sicau.edu.cn. Linyuan Shen, Email: shenlinyuan@sicau.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The raw sequencing data of tRF and tiRNA sequencing was deposited in the National Genomics Data Center accession PRJNA899798 and 10.3390/ani12243561. Any additional raw data that backs up the findings of this study is available from the corresponding author upon request.