A novel microprotein MUCP1 promotes colorectal cancer metabolic reprogramming by regulating mitochondrial succinate transport

Abstract

Background

Metabolic reprogramming is a hallmark of colorectal cancer (CRC), yet the molecular regulators that orchestrate this process remain incompletely understood. Although many long non-coding RNAs (lncRNAs) possess protein-coding potential, their translational products and metabolic functions have been largely overlooked. Here, we identify MUCP1, a microprotein encoded by the lncRNA MUC20-OT1, as a critical regulator of mitochondrial metabolism and epigenetic remodeling in CRC.

Methods

Multi-omics data were integrated to identify MUC20-OT1 as a candidate lncRNA encoding a functional microprotein. Fusion reporter plasmids, mass spectrometry, and immunoblotting were used to validate MUCP1 translation and mitochondrial localization. Functional assays, metabolomic profiling, ¹³C₅-glutamine isotope tracing, subcellular succinate quantification, CUT&Tag, and xenograft models were performed to investigate the role of MUCP1 in facilitating mitochondrial succinate export and maintaining glutamine metabolism homeostasis.

Results

The microprotein MUCP1, encoded by the lncRNA MUC20-OT1, serves as an auxiliary regulator of SLC25A10-mediated mitochondrial succinate transport. MUCP1 is upregulated during CRC progression and localizes in the mitochondrial outer membrane, where it facilitates the balance of mitochondrial succinate metabolism. Elevated extramitochondrial succinate subsequently enhances H3K4me3 histone modifications, promoting the transcription of enzymes involved in glutamine metabolism and sustaining the high metabolic demands of CRC cells.

Conclusions

This study identifies MUCP1 as a novel lncRNA-encoded microprotein that maintains metabolic homeostasis in CRC by coupling mitochondrial succinate transport to histone methylation. MUCP1 might be a promising metabolic vulnerability and therapeutic target in CRC.

Graphical abstract

The lncRNA-derived microprotein MUCP1 coordinates succinate export and epigenetic activation of glutamine metabolism to sustain CRC metabolic reprogramming.

Introduction

Colorectal cancer (CRC) ranks as the third most prevalent cancer globally, with a steadily increasing mortality rate [1,2]. The primary contributor to the high mortality rate in CRC patients is metastasis, with a significant proportion of newly diagnosed patients already presenting with metastatic disease [3,4]. Cancer cells adapt to various metabolic demands required for their progression by utilizing a range of metabolites, including glucose, glutamine, and fatty acids [[5], [6], [7]]. However, there remains a lack of comprehensive understanding of the complexity of cellular metabolic activities during CRC progression. Investigating the dynamic changes in metabolites throughout CRC progression is crucial for accelerating the development of effective metabolic therapies.

Historically, non-coding RNAs (ncRNAs) were dismissed as non-functional transcriptional noise [[8], [9], [10]]. However, long noncoding RNAs (lncRNAs) have emerged as pivotal players in CRC progression, significantly impacting metabolic reprogramming [[11], [12], [13]]. Intriguingly, mounting evidence suggests that several lncRNAs, initially misannotated as non-coding, in fact encode peptides or proteins [[14], [15], [16]]. These novel lncRNA-derived products, such as those from HOXB-AS3 [17], LINC00467 [18], LINC00116 [19], and LINC00266 [20], are crucial for regulating essential metabolic activities, including glucose metabolism, fatty acid oxidation, and ATP production, thereby modulating various malignant characteristics of CRC. This discovery opens new avenues for exploring the complex interplay between tumor metabolism and CRC malignancy. Nonetheless, the exact functions of many lncRNA-derived proteins or peptides in CRC progression remain elusive, highlighting the need for further detailed studies.

In this study, we identify a novel microprotein, MUCP1, encoded by the lncRNA MUC20-OT1, which is markedly overexpressed in CRC. MUCP1 facilitates SLC25A10-mediated succinate transport from mitochondria and epigenetically activates genes involved in glutamine metabolism, supporting the elevated metabolic demands of CRC. Succinate not only promotes epithelial–mesenchymal transition (EMT) [21,22] and tumor-associated macrophage polarization, but also inhibits histone and DNA demethylases, thereby reshaping the epigenetic landscape [[23], [24], [25]]. Despite its recognized oncogenic roles, the regulatory mechanisms controlling succinate transport in cancer cells remain elusive. Our findings identify MUCP1 as a functional microprotein and a potential therapeutic target in the context of CRC metabolic reprogramming.

Materials and methods

Cell culture

Human colorectal cells HCT 8, HCT 15, HCT 116, SW620, SW480, Lovo, FHC, human liver cells Hep G2 and Hep3b, human breast cells MCF10A, MB-231, MCF7, and T-47D, along with HEK293T cells, were procured from the ATCC (USA) or Chinese academy of science Cell bank (Shanghai, China) and cultured under standard conditions. In brief, the above-mentioned cells were cultured in their respective specialized media containing 10% fetal bovine serum and antibiotics at 37°C in a 5% CO₂ atmosphere.

Plasmid constructs and transfection

The pcDNA3.1(+) vector was procured from GenScript Company (Nanjing, China). Mutant plasmids comprising GFPmut, ORF-GFPmut, and ORF-FLAG were fabricated within the customer service provided by GenScript. In ORF-FLAG plasmid, three consecutive FLAG sequences were tandemly fused to the C-terminal end of MUCP1 open reading frame. The lentivirus expressing shRNA targeting LncRNA MUC20-OT1 and lentivirus selectively expressing MUCP1-3 × FLAG, both of which were synthesized by GeneChem Co., Ltd. (Shanghai, China). Three knockdown target sites were designed for the non-open reading frame sequence of MUC20-OT1. A vector was constructed with these three targets linked together and integrated into a lentiviral shRNA vector containing the U6 promoter. The specific shRNA sequences are provided in Table S1. Following infection, cells were cultured for 72 hours and subjected to selection in either puromycin (2 μg/mL) or hygromycin (100 mg/mL). The efficiency of knockout and protein expression was evaluated using fluorescence microscopy, qPCR and Western blot analysis.

CRISPR/Cas9-mediated endogenous FLAG knock-in

To generate the MUCP1-3xFLAG knock-in cell line, an sgRNA was designed to target the stop codon region of the MUCP1 open reading frame to facilitate the C-terminal integration of a 3xFLAG epitope. The knock-in strategy was performed via microhomology-mediated end-joining, as previously described[26,27]. The specific sgRNA sequence (synthesized by Hanbio, Shanghai, China) is listed in supplementary file. Briefly, cells were transfected with the expression vector and a dsDNA donor template via standard transfection reagents. Positive cells were selected with puromycin to establish a stable knock-in pool.

RNA interference

Cells were transfected with siRNA against the SLC25A10 gene (50 nM) or NC siRNA using the provided riboFECT transfection reagent for 48 hours. Antisense Oligonucleotides (ASO) targeting the MUCP1 (75 nM) were applied for varying durations based on experimental requirements, with ASO target design referencing shRNA and not targeting the open reading frame sequence of MUCP1. The siRNA, ASO, and transfection reagents were manufactured and supplied by RiboBio (Guangzhou, China), with the sequences of siRNA and ASO listed in Table S1.

Criteria for selecting anti-MUCP1 antibodies

Given the high sequence homology between MUCP1 and human SDHA proteins, despite their markedly different molecular weights, and the availability of well-characterized SDHA antibodies, we employed SDHA antibodies to detect endogenous MUCP1. The selected antibodies recognize peptide sequences shared by both proteins—one located at the N-terminus and the other within the mid-region of MUCP1. A schematic representation of the epitope locations is provided in Figure S2.

RNA extraction, qRT-PCR and bulk RNA-seq analysis

Expression levels of genes including MUC20-OT1 and SLC25A10 were evaluated using qRT-PCR under various experimental conditions. Total RNA was isolated from designated cells using RNA-easy Isolation Reagent (Vazyme) and then reverse transcribed into cDNA using the HiScript Reverse Transcription Kit (Vazyme). Quantification was performed using the QuantStudio Q5 System (Applied Biosystems, USA). Gene expression levels were quantified using the 2-ΔΔCt method, with β-actin used as the normalization control for data normalization. Primers used for detecting the above genes were synthesized by Ribobio (Guangzhou, China). The Real-time PCR experiments were conducted independently and repeated three times. Primer sequences for the genes involved are provided Table S2.

The extracted RNA samples underwent strict quality control, primarily quantified using Nanodrop and assessing RNA integrity using Agilent 4200 TapeStation, followed by cDNA library construction. Illumina sequencing was performed on the library samples using the Illumina NovaSeq 6000 sequencing platform with PE150 mode. The reads filtered from each sample were aligned to the reference sequence (HUMAN GENCODE 34 GRCh37 (11.2019)) using star software. Quantitative analysis was performed using featureCounts software, with gene expression quantified using TPM (Transcripts Per kilobase of exon model per Million mapped reads) as the unit. Differentially expressed genes were identified using edgeR package, with a filtering threshold set at FDR< 0.05 and |log2(FoldChange)| > 1, and the results were used for subsequent analysis.

Sodium carbonate extraction of mitochondrial proteins

Mitochondria were isolated from cells according to the manufacturer’s instructions (Thermo Fisher, USA). To determine the submitochondrial localization of the MUCP1 protein, the mitochondrial pellet was resuspended in 100 mM Na₂CO₃ and incubated on ice for 30 min to facilitate protein extraction. The suspension was then subjected to ultracentrifugation at 125,000 × g for 30 min at 4°C, yielding a soluble supernatant fraction and a membrane-enriched pellet fraction. Both the supernatant and pellet fractions were subsequently collected and analyzed by Western blotting.

Western blot analysis

Cells were suspended in RIPA lysis buffer and then subjected to high-temperature denaturation at 100 °C for 8 min. The cell lysate was centrifuged at 12000 rpm for 15 min. Protein concentration was determined using the BCA Protein Concentration Assay Kit (Beyotime, Shanghai). To detect MUCP1, MUCP1-GFP, or MUCP1-Flag peptides, whole cell lysates and tissue lysates were separated by 12 % Tris-SDS-PAGE and then transferred onto 0.22 μm polyvinylidene difluoride membranes (Millipore, USA) by electroblotting. These membranes were incubated overnight at 4 °C with antibodies against GFP (1:2000), Flag (1:2000), or MUCP1 (1:1000), along with Tubulin antibody (1:5000) as a reference. Subsequently, the membranes were incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (1:10000) at room temperature for 2 h. The signal was detected using an Enhanced Chemiluminescence kit (NCM Biotech, Suzhou) and imaged with a chemiluminescence imaging system (Tanon, Shanghai).

Immunofluorescence staining

For subcellular localization, HCT8 cells stably expressing MUCP-FLAG were seeded onto glass coverslips in 24-well plates at a density of 1 × 10^4 cells per well. Cells were first stained with 200 nM MitoTracker for 20 min at 37 °C, then fixed in 4 % PFA and permeabilized with 0.1 % Triton X-100 at room temperature (RT). Following a 2 h blocking step with 5 % BSA at 4 °C, samples were incubated with anti-FLAG (1:200, 2 h, RT) and corresponding secondary antibodies (1:500, 1 h, RT). Nuclei were visualized via DAPI staining. Representative images were acquired using Zeiss Axio Iamgera.

Migration and invasion assays

In vitro migration and invasion assays were performed using Transwell chambers placed in 24-well plates, with each chamber coated with 50 μg of Matrigel gel (1:8 dilution). The upper chamber was filled with serum-free cell suspension at a concentration of 5 × 10^5 cells/ml, with a total volume of 200 μl. RPMI-1640 medium and 10 % FBS were added to the 24-well plate (no substrate gel coating was necessary for migration assays). After the incubation period, non-migrated cells on the chamber surface were gently removed with a cotton swab. The remaining cells were fixed with paraformaldehyde and stained with 5 % crystal violet. Images were acquired using an Olympus FSX100 microscope (Olympus, Japan), and cells on the lower surface of the membrane were counted.

Cellular growth assay

CRC cells were transfected with ASO for 24 h. Cell proliferation assay was conducted using CCK-8. Approximately 2000 cells were seeded per well in a 96-well plate. After overnight stabilization of the cells, CCK-8 reagent (10 μl per well) was added to the culture wells for continuous incubation for 2 h each day for five consecutive days. Absorbance at 450 nm wavelength was then measured using a 96-well plate reader, with absorbance values directly proportional to the number of proliferating cells. For colony formation assays, cells treated as described above were seeded into six-well plates at a density of 500 cells per well and cultured for two weeks. Cells were fixed with paraformaldehyde for 15 min, washed, stained with crystal violet staining solution, and counted.

RNA fluorescence in situ hybridization (FISH)

Cy3-labeled probes targeting LncRNA MUC20-OT1, 18S, and U6 were designed and synthesized by Ribobio (Guangzhou). FISH assays were performed according to the manufacturer's protocol using the FISH Kit from RiboBio.

Identification of endogenous MUCP1 by mass spectrometry

To identify endogenous MUCP1, total protein from HCT8 cells was separated by 12 % Tris-SDS PAGE and silver-stained. Gel bands below approximately 20 kDa were excised and digested with trypsin. Protein identification was performed using MS method as described previously [20]. Mass spectrometry detection was conducted by Gene-denovo Company (Guangzhou).

Following enzymatic digestion, peptides were sequentially extracted from gel pieces using 50 % acetonitrile solution containing 0.1 % formic acid, 80 % acetonitrile solution containing 0.1% formic acid, and pure acetonitrile. The peptides were reconstituted in solvent A (A: 0.1 % formic acid in water) and analyzed using an Orbitrap Fusion coupled to an EASY-nanoLC 1200 system (Thermo Fisher Scientific, MA, USA). A 3 μL peptide sample was loaded onto a 25 cm analytical column (75 μm inner diameter, 1.9 μm resin) and separated using a 60 min gradient starting at 4 % buffer B (80 % ACN with 0.1 % FA) for 6 min, followed by a stepwise increase to 28 % over 40 min, 90 % over 5 min, and held for 15 min. The column flow rate was maintained at 300 nL/min with a column temperature of 40 °C. The electrospray voltage was set to 2 kV. The mass spectrometer operated in data-dependent acquisition mode, automatically switching between MS and MS/MS acquisition. Finally, tandem mass spectra were analyzed using PEAKS Studio version 10.6 (Bioinformatics Solutions Inc., Waterloo, Canada).

HPIC-MS/MS metabolomics analysis

The alterations in metabolites during the metabolic remodeling process of CRC cells mediated by MUCP1 were investigated through untargeted metabolomics and targeted metabolomics analyses. Metabolite analysis was performed by Personalbio Company (Shang hai). In brief, HCT8 cells cultured in 10 cm² dishes for 24 h along with the culture supernatant (approximately 1 × 10^7 cells per sample) were simultaneously collected. The separated culture supernatant was centrifuged at 12000 g, 4 °C for 10 min to remove cell debris and other impurities. The cells were washed twice with cold PBS and then dissolved in 800 μL cold methanol/acetonitrile (1:1, v/v) to remove proteins and extract metabolites. The mixture was collected in new centrifuge tubes and centrifuged at 14000 g for 20 min to collect the supernatant.

For metabolites extraction, adding 500 μL precooled MeOH/H2O (3/1, v/v), the samples were vortexed for 30 s. The samples were precooled in dry ice, repeated freeze-thaw three times in liquid nitrogen, followed by vortexed for 30 s and sonicated for 15 min in ice-water bath. After incubating at -40 °C for 1 h, the samples were centrifuged at 12000 rpm (RCF = 13800(× g), R = 8.6 cm) and 4 °C for 15 min. A 400 μL aliquot of the clear supernatant was collected and dried by spin. Then the samples were reconstructed with 200 μL of purified water. Reconstituted samples were vortexed before filtration through the centrifuge tube filter and were subsequently transferred to inserts in injection vials for HPIC-MS/MS analysis. This experiment utilized the Thermo Scientific Dionex ICS-6000 HPIC (Thermo Scientific) for high-performance ion chromatography. The protective column employed was AG11-HC RFIC, 2 × 50 mm, while the separation column was AS11-HC RFIC, 2 × 250 mm. Subsequently, metabolite analysis was conducted using the 6500 QTrap+ mass spectrometer in multiple reaction monitoring (MRM) mode.

Metabolic flux of tricarboxylic acid cycle (TCA) cycle derived from glutamine

Steady metabolic flux analysis was performed as previously described [28]. For glutamine labeling, cells were cultured in RPMI-1640 without glutamine, supplemented with 10 % fetal bovine serum, and 2 mM ¹³C₅-glutamine for 24 h to achieve steady state. After two washes with cold PBS, 1 ml of 4:1 (v/v) methanol-water pre-cooled at -80 °C for 4-6 h was added to the dishes and incubated at -80 °C for 20 min. Cells were scraped off with a cell scraper, ensuring a cell count >1 × 10⁷ cells per sample, and sent to LipidALL Technologies (Changzhou, China) for metabolic flux analysis.

Mitochondrial isolation and quantification of succinate, glutamine, and glutathione

In brief, CRC cells transfected with ASO or subjected to other treatments were harvested along with culture supernatants at specified time points. To measure the levels of succinate in mitochondria and extracellular components, cell mitochondria were isolated using a commercial mitochondrial extraction kit (Thermo Scientific). To help prevent rapid succinate leakage, diethyl butylmalonate (DEBM) was included during isolation. Protein concentrations across samples were measured using the BCA assay. Succinate, glutamine, and glutathione levels in the collected samples were determined using succinate assay kit (Sigma Aldrich), glutamine assay kit (Solarbio), and glutathione assay kit (KeyGen Biotech), respectively, according to the manufacturer's instructions, and adjusted based on the protein concentration. In brief, lyse 5 × 10⁶ cells as per the experimental objectives, separate the required cellular components, and centrifuge at 4 °C. Collect the supernatant, measure the absorbance following the manufacturer's instructions, and calculate the metabolite concentrations. For the calculation of cellular glutamine uptake, the glutamine level in fresh complete medium is measured and subtracted from the glutamine concentration in the supernatant of cultured cells collected at designated time points. This difference indicates the glutamine uptake by cells under different conditions.

CUT&TAG

CRC cells treated with DES or DMSO for the specified durations underwent CUT&Tag, with 50,000 cells per sample. Following the manufacturer's instructions, DNA fragments modified with H3K4me3 were extracted from cells using the CUT&Tag Assay Kit (pAG-Tn5) for Illumina. After DNA recovery from cells, library construction was directly performed through PCR amplification, and the libraries were sent to Personalbio Company (Shang hai) for sequencing.

Chromatin immunoprecipitation-qPCR assay

The chromatin immunoprecipitation (ChIP) assay was conducted using the Sonication ChIP Kit (RK20258, ABclonal, Wuhan) according to the provided protocol. Briefly, Cells (2 × 10^7 cells) were fixed in formaldehyde and sonicated to shear DNA into 200-500 bp fragments. Chromatin was then immunoprecipitated with H3K4me3 antibodies or normal mouse IgG for 5 h at 4 °C, followed by a 2 h incubation with Protein A magnetic beads. After proteinase K treatment and cross-link reversal, DNA was extracted via a spin column-based method (RK30100, ABclonal, Wuhan). The resulting DNA template was resuspended in ddH2O and subjected to qPCR for validation.

Animal model

In accordance with the approval from the Animal Care Committee of Nanjing Medical University, five-week-old female BALB/c nude mice were maintained under pathogen-free conditions. HCT 8 or HCT 15 cells (5 × 10⁶ /0.1 ml PBS) were implanted into the axilla of each mouse. Tumor growth was monitored every 3-4 days using digital calipers, and tumor volume was calculated using the formula: Volume = 1/2 (length × width^2). Tumor growth was monitored every 3-4 days using digital calipers, and tumor volume [V = (L * W²)/2] was calculated as previously described [29]. After twenty-two days, the mice were euthanized, and tumor volumes were measured. Excised tumor samples were paraffin-embedded for immunohistochemical staining.

Immunohistochemistry staining

Briefly, tumor tissues were fixed in formalin, paraffin-embedded, sectioned into 4 μm slices, and then processed using standard deparaffinization and rehydration techniques. Immunodetection was conducted using anti-H3K4me3 (1:200), anti-Ki67 (1:1000), anti-SLC25A10 (1:200), and anti-E-cadherin (1:500).

Quantification and statistical analysis

Statistical analysis was carried out using GraphPad Prism 9.0. For in vitro and in vivo experiments, Two-tailed student’s test was employed to evaluate statistical distinctions between different groups. Statistical significance was associated with P values <0.05 and N.S. for no statistical significance. All data are presented as the mean ± standard deviation from at least three independent replicates. Other materials and methods are provided in the supplementary methods file.

Results

Identification of coding-capable lncRNAs in CRC

To identify lncRNAs encoding functional microproteins associated with CRC metabolism, we conducted an integrative analysis of ribosome profiling data retrieved from the GEO (GSE133925, GSE143263, and GSE129504) database [[30], [31], [32]]. LncRNAs harboring the following attributes are prime candidates for assessing coding capacity: (1) high phyloCSF [33] and CPAT scores [34], resembling predictive features of coding proteins, (2) highly conserved amino acid sequences with known proteins, and (3) predicted functions of lncRNAs are linked to tumor metabolism (Fig. 1A). The results indicate that 16 ribosome-bound lncRNAs are shared among the three datasets. The coding potential of these lncRNAs was further assessed using phyloCSF and CPAT algorithms. GAPDH gene served as a positive control, while lncRNA XIST served as a negative control, being a known non-coding gene. Moreover, LINC00998 was chosen as a reference due to its recent discovery of producing functional peptides that activate the MAPK pathway by interacting with tyrosine kinases SRC/YES1 [30]. The results revealed that only MUC20-OT1 (LINC00969) obtained reliable scores in both algorithms (Fig. S1A, S1B). The ORF finder program from NCBI tools was utilized to predict the amino acid sequence encoded by candidate lncRNAs and to perform protein homology and conservation analysis. Notably, sequence comparisons revealed a high similarity between the potential protein encoded by MUC20-OT1 and the succinate dehydrogenase flavoprotein subunit (SDHA) across multiple species, with a match of 92.36% in humans, suggesting that MUC20-OT1 may encode a unidentified protein (Fig. 1B, S1C). As the core catalytic subunit of succinate dehydrogenase, SDHA catalyzes the oxidation of succinate to fumarate within the TCA cycle [35,36]. Genetic alterations or functional impairments in SDHA are frequently linked to tumorigenesis [35,37,38]. Supporting this metabolic relevance, AnnoLnc2 database [39] analysis revealed that MUC20-OT1 is functionally enriched in pathways related to cellular metabolism and MAPK signaling (Fig. S1D).

Fig. 1.

Identification and validation of MUCP1 as a microprotein encoded by lncRNA MUC20-OT1.

(A) Diagram depicting the selection process for lncRNAs according to established criteria.

(B) Comparative analysis of MUCP1 amino acid sequences across different species. Per. Ident, percentage of identity. Acc. Len, access length.

(C) Schematic of the GFPmut fusion plasmid used for transfection. The wild-type GFP start codon (ATGGTG) was mutated to ATTGTT (GFPmut), and the predicted ORF1/2 start codon (ATG) was mutated to ATT.

(D) Western blot analysis of ORF1/2-GFPmut fusion protein expression in 293T cells 48 hours post-transfection.

(E) Expression of the ORF1/2-GFPmut fusion protein was assessed by Western blot after mutating the start codon of ORF1/2.

(F) Diagram of the constructed FLAG fusion plasmid, in which a triple FLAG tag sequence was fused in tandem to the C-terminus of ORF1.

(G) Following the transfection of the plasmid into 293T cells for 48 hours, Western blot analysis was conducted to detect ORF1-FLAG fusion proteins.

(H) Schematic illustration showing the insertion of a triple FLAG sequence at the C-terminus of the ORF1 locus via CRISPR/Cas9. Successful knock-in was verified by Western blot using an anti-FLAG antibody.

Additionally, we utilized Gene Expression Profiling Interactive Analysis (GEPIA) databases to analyze prognostic significance of MUC20-OT1 in colorectal cancer. The Kaplan-Meier curves from the GEPIA database indicated a significant association between the expression level of MUC20-OT1 and the overall survival of CRC patients (Fig. S1E). These above results indicate that MUC20-OT1 binds to active ribosomes, potentially encoding a protein involved in tumor metabolism. Thus, in this study, we primarily focus on MUC20-OT1 as a candidate encoding lncRNA.

lncRNA MUC20-OT1 encodes a microprotein

To assess the protein-coding potential of MUC20-OT1, we utilized ORF Finder to predict putative open reading frames (ORFs) and designated the corresponding protein product as MUCP1. To experimentally validate the protein-coding potential of MUC20-OT1, we constructed GFP-tagged fusion plasmids by fusing the predicted ORFs to the C-terminus of GFP. The canonical ATGGTG start codon of GFP was mutated to ATTGTT to ensure that translation initiation originated solely from the candidate ORFs (Fig. 1C). Two putative ORFs, ORF1 and ORF2, were identified via ORF Finder, both sharing an identical N-terminal amino acid sequence of the first 39 residues (Table S3). Commercially available antibodies targeting this shared N-terminal region enabled precise detection of translation products (Fig. S2A). Both ORF1-GFPmut and ORF2-GFPmut fusion proteins were expressed, with observed sizes matching predicted molecular weights (Fig. 1D). Mutating the start codon within the MUCP1 ORFs abolished fusion protein expression (Fig. 1E). Importantly, we detected an endogenous band around 15 kDa, aligning with the predicted size of ORF1, which lends further support to the notion that MUC20-OT1 may encode a functional microprotein. Based on the cumulative evidence, we focused our subsequent investigations on ORF1. To verify whether ORF1 is independently translated, we constructed an ORF1-3 × FLAG fusion plasmid (Fig. 1F). Robust expression of the FLAG-tagged protein was observed, whereas mutation of the start codon completely abolished protein expression (Fig. 1G). Furthermore, endogenous in-frame knock-in (KI) of an epitope tag at the C-terminus of ORF1 confirmed the production of the microprotein via immunoblot (Fig. 1H). Collectively, these findings suggest that MUC20-OT1 gives rise to a functional microprotein, referred to here as MUCP1.

MUCP1 is an endogenously outer mitochondrial membrane protein

To further validate the authenticity of MUCP1 expression, we performed both overexpression and knockdown experiments targeting the MUC20-OT1 using plasmid-based overexpression and antisense oligonucleotides (ASO), respectively. MUCP1 levels significantly increased following MUC20-OT1 overexpression (Fig. 2A), while silencing MUC20-OT1 markedly suppressed MUCP1 expression (Fig. 2B, C). In contrast, SDHA expression remained unaffected by these manipulations. To exclude the possibility that MUCP1 is a proteolytic product of its homolog SDHA, we knocked down SDHA and observed no changes in MUCP1 expression (Fig. 2D). Finally, tandem mass spectrometry analysis of the <20 kDa protein fraction identified MUCP1-specific peptides, further supporting its classification as a distinct, independently translated microprotein (Fig. S2B). To assess the expression of endogenous MUCP1 in CRC, we extracted total RNA and protein from CRC tissues and cell lines. qRT-PCR analysis revealed that transcripts encoding the MUCP1 isoform of MUC20-OT1 were significantly upregulated in CRC cell lines compared to normal colonic epithelial cells, except in SW480 cells where no significant change was observed (Fig. 2E). In addition, RNA-FISH analysis revealed that MUC20-OT1 is abundantly localized in both the cytoplasm and nucleus of CRC cells (Fig. S2C). The endogenous presence of MUCP1 was further validated in colorectal, breast, and liver cancer cell lines (Fig. 2F, S2D). Moreover, analysis of paired fresh CRC and adjacent normal tissues revealed consistently elevated MUCP1 expression in tumor samples (Fig. 2G).

Fig. 2.

MUCP1 is a novel protein located on the mitochondrial outer membrane and is widely expressed across multiple tissues.

(A) Expression levels of MUCP1 and SDHA were analyzed in 293T cells 48 hours after transfection with a plasmid overexpressing full-length MUC20-OT1.

(B) MUC20-OT1 mRNA expression was analyzed in HCT8 and HCT15 cells after transfection with control or MUCP1-targeting ASO.

(C) MUCP1 and SDHA expressions were analyzed in CRC cells after transfection with control or MUCP1-targeting ASO.

(D) The expression levels of MUCP1 and SDHA were evaluated in CRC cells after transfection with either control siRNA or si-SDHA.

(E) Relative expression of MUC20-OT1 in various CRC cell lines and normal colonic epithelial FHC cells.

(F) Detection of MUCP1 in CRC, liver cancer, and breast cancer cells using an antibody homologous to SDHA (113-128 aa).

(G) Assessment of MUCP1 levels in the indicated CRC tumor (T) and adjacent non-tumor (NT) tissues.

(H) Isolation of nuclei and mitochondria from SW480 and HCT8 cells to examine MUCP1 expression in different cellular compartments. Lamin B1 and Cox IV serve as markers for the nucleus and mitochondrial matrix, respectively.

(I) Representative immunofluorescence images showing the subcellular localization of MUCP1-FLAG in HCT8 cells. Cells were stained with anti-FLAG antibody (green) to detect MUCP1 expression and Mito-Tracker (red) to visualize mitochondria. Nuclei were counterstained with DAPI (blue). The merged image highlights the colocalization of MUCP1 with mitochondria.

(J) Signal IP-4.1 predicted results of MUCP1. Data are presented as the mean ± standard deviation. Statistical significance was assessed using a two-sided unpaired Student’s T-test.

To further characterize this microprotein, we conducted subcellular fractionation, revealing its remarkable distribution in mitochondria (Fig. 2H). Supporting this, MUCP1-FLAG fusion protein was observed within mitochondria via immunofluorescence staining (Fig. 2I). However, analysis using SignalP 4.1 [40] indicated that MUCP1 lacks a canonical mitochondrial targeting sequence, suggesting that it may associate with the mitochondria peripherally rather than translocate into the mitochondrial interior (Fig. 2J). Proteinase K-based mitochondrial topology assays further support the localization of MUCP1 to the outer mitochondrial membrane (Fig. S2E). To further define its membrane association, we performed sodium carbonate (Na₂CO₃) extraction on isolated mitochondria[41]. MUCP1 was primarily detected in the pellet fraction, identifying it as an integral membrane protein. Consistently, the outer membrane marker TOMM20 partitioned into the pellet, while the soluble mitochondrial protein HSPA9 was recovered in the supernatant (Fig. S2F). Taken together, our findings suggest that MUCP1, encoded by MUC20-OT1, is a conserved microprotein closely related to mitochondria.

MUCP1 functions as an oncogenic microprotein to maintain CRC progression

We proceeded to investigate the potential role of upregulated MUCP1 in CRC progression. The silence of MUC20-OT1 impeded the growth, migration, and invasion of CRC cells. Rescue assays were executed by transfecting CRC cells with plasmids encoding FLAG-tagged MUCP1 and a mutant MUCP1 (MUCP1mut) featuring a mutated start codon, against a backdrop of diminished MUC20-OT1 expression. The transfection of MUCP1-FLAG plasmids significantly reinstated the proliferative, invasive, and migratory functions of the ASO-treated cells, a restoration not observed with the MUCP1mut-FLAG plasmids, as they showed no significant changes compared to the MUCP1-FLAG group (Fig. S3A, 3B). Additionally, the efficiency of MUCP1 knockdown in CRC cells using shRNA is depicted in Figure S3C and D. The in vivo growth of xenograft tumors derived from cancer cells with stable MUCP1 knockdown was markedly inhibited (Fig. S3E, S3F). Collectively, these findings suggest that the microprotein MUCP1, encoded by the lncRNA MUC20-OT1, rather than the lncRNA itself, contributes to tumor progression both in vitro and in vivo, supporting a potential oncogenic role in CRC.

Reduced MUCP1 disrupts metabolic homeostasis in CRC cells

Mitochondria, the cellular hubs for energy metabolism, exhibits metabolic remodeling in cancer cells compared to normal cells [42,43]. Considering that MUCP1 is an uncharacterized protein located on the mitochondrial outer membrane, it leads us to hypothesize that MUCP1 has a potential role in the metabolic processes in CRC. We further employed untargeted metabolomics to analyze the metabolites in cellular contents of CRC cells silenced for MUCP1. The results revealed a notable decrease in glutamine and its derivatives such as DL-Glutamic acid, as well as amino acids critical for tumor growth including threonine and asparagine (Fig. 3A, left). Additionally, it was evident that compared to control cells, shMUCP1 cells showed a significant increase in the absorption of glutathione, arginine, and aspartate from the culture medium (Fig. 3A, right). Tumor cells often exhibit increased dependence on glutamine, which is essential for the biosynthesis of various key metabolites such as amino acids (glutamate, etc.), TCA cycle intermediates (alpha-ketoglutarate, succinate, etc.) [44,45], and glutathione, which mediates antioxidative responses [46]. Considering the close link between MUCP1 and succinate metabolism, along with the Warburg effect [6], which shifts glutamine catabolism toward α-ketoglutarate and succinate production for TCA cycle replenishment, we turned our attention to analyzing key TCA cycle intermediates.

Fig. 3.

The oncogenic protein MUCP1 is involved in succinate and glutamine metabolism

(A) Intracellular and culture supernatant metabolites were compared between MUCP1-deficient and control cells using untargeted metabolomics (n = 5). The figure displays those metabolites showing statistically significant differences.

(B) Abundance of glycolysis and tricarboxylic acid cycle pathway metabolites in indicated cells. Data derived from targeted metabolomics (n = 5).

(C) Top 20 enriched KEGG pathways of differentially expressed genes in HCT8 cells with silenced MUCP1.

(D) Western blot analysis of indicated proteins in the MAPK pathway in cells with silenced MUCP1 compared to control cells.

(E) Western blot analysis of indicated proteins in the PI3K-AKT pathway.

(F-I) Relative concentrations of glutamine (F), glutathione (G), and succinate (H) in shMUCP1 and control cells. Intracellular metabolite analysis was performed using equal numbers of cells. Succinate concentrations in the culture medium were measured after 24 hours of incubation (I).

Data are presented as the mean ± SD. Statistical significance was assessed using a two-sided unpaired Student's t-test.

Through more precise targeted metabolic strategies, we analyzed the metabolic characteristics of cells deficient and rich in MUCP1. It is noteworthy that cells with silenced MUCP1 demonstrated elevated levels of alpha-KG and pyruvate, whereas succinate levels were significantly lower compared to controls, with unchanged levels of fumarate, indicating a bottleneck at succinate in the TCA cycle (Fig. 3B). High-throughput RNA sequencing was performed on MUCP1-silenced cells. KEGG pathway analysis of differentially expressed genes showed significant enrichment in the MAPK signaling pathway and various hormone secretion and release pathways, as depicted in Fig. 3C. The PI3K/AKT [47] and MAPK pathways, previously identified as over activated in cancer, facilitate metabolic processes like glycolysis and glutamine utilization to fulfill the metabolic requirements [48,49]. We further explored the impact of MUCP1 silencing on the MAPK pathway, observing decreased phosphorylation levels of ERK and p38 via Western blotting. Similarly, the activity of the PI3K-AKT signaling pathway was also diminished (Fig. 3D, E). To ascertain the effects of MUCP1 on glutamine and succinate metabolism, we assessed the levels of glutamine (Fig. 3F), glutathione (Fig. 3G), succinate (Fig. 3H), and extracellular succinate (Fig. 3I) in CRC cells with silenced MUCP1 via colorimetric methods. The results were consistent with those from metabolomic data. Overall, these findings indicate that MUCP1 deficiency results in lowered metabolic activity, possibly due to disrupted glutamine metabolism and an imbalance in succinate homeostasis.

MUCP1 deficiency suppresses glutamine utilization via reduced succinate-driven H3K4me3 modification

To clarify how MUCP1 regulates metabolic homeostasis during CRC progression, we investigated the cause of reduced intracellular glutamine levels in MUCP1-deficient cells. Interestingly, MUCP1 knockdown suppressed CRC progression (Fig. S3A-F), yet paradoxically resulted in elevated consumption of glutathione and several amino acids from the culture medium (Fig. 3A, right). To determine whether the elevated extracellular glutathione uptake was due to compromised intracellular glutamine metabolism and reduced endogenous glutathione production, we measured the mRNA levels of GOT1, GLUL, GLUD1, and GLS1, enzymes involved in glutamine catabolism. As depicted in Fig. 4A, suppression of MUCP1 expression led to a general downregulation of key genes in the glutamine metabolism, with GLUD1 and GLS1 being particularly affected. Given the potential role of MUCP1 in succinate metabolism, we were intrigued to find that this inhibitory effect was largely reversed by supplementation with diethyl succinate (DES), a membrane-permeable succinate analog. Multiple studies have shown that succinate accumulation, driven by MYC activation [50], PTEN deficiency [51], or impaired SDH function [25], activates H3K4me3 and tumor-specific gene expression, thereby promoting tumor progression [52]. Therefore, we hypothesize that MUCP1-mediated reduction in succinate levels may lead to decreased H3K4me3, thereby impairing glutamine utilization in CRC cells. Consistent with this, MUCP1 silencing led to a significant reduction in H3K4me3 levels (Fig. 4B) and a marked increase in the α-KG/succinate ratio (Fig. S4A), which functions as a metabolic sensor of the cellular epigenetic landscape, impacting both DNA and histone methylation [53]. Furthermore, we identified numerous H3K4me3 modification sites in the promoter regions upstream of 1000 bp from the transcriptional start site (TSS) of the four key glutamine-related genes and MUC20-OT1 (Fig. S4B). To validate the central regulatory role of H3K4me3 in the transcriptional activation of these genes, we employed the small-molecule inhibitor OICR-9429, which blocks H3K4me3 formation by disrupting histone methyltransferase activity[54]. Our results showed that while DES significantly up-regulated the expression of glutamine metabolism genes and MUC20-OT1, this effect was markedly attenuated by OICR-9429 (Fig. S4C, S4D). Further CUT&Tag analysis confirmed a reduction in H3K4me3 peaks at the TSS of these genes in shMUCP1 cells, whereas DES supplementation significantly restored global H3K4me3 levels (Fig. 4C, D). Consistently, ChIP-qPCR analysis corroborated that MUCP1 knockdown reduced H3K4me3 enrichment at the GLS1, GLUL, and MUC20-OT1 promoters, which was significantly rescued by DES or KDOAM-25 [55,56], a KDM5 inhibitor that enhances H3K4me3 occupancy (Fig. S4E-G). Our results imply that the decline in H3K4me3 enrichment at the promoters of key glutamine metabolic genes may contribute to the reduced glutamine availability and metabolic dysfunction observed in MUCP1-deficient cells.

Fig. 4.

MUCP1 inhibition reduces genomic H3K4me3 levels and disrupts glutamine metabolic activity.

(A) The relative mRNA levels of key glutamine metabolism genes (GOT1, GLUL, GLUD1, and GLS1) in HCT8 and HCT15 cells under the indicated conditions. An unpaired two-tailed t-test was used to determine the significance of differences in gene expression levels between groups.

(B) Western blot analysis of H3K4me3 levels in HCT8 and HCT15 cells with MUCP1 knockdown, showing reduced H3K4me3 levels compared to control.

(C) CUT&Tag profiles illustrating H3K4me3 peak density across ± 3 kb regions flanking the transcription start and end sites of the indicated genes under specified conditions.

(D) Analysis of H3K4me3 enrichment at promoter regions of MUC20-OT1 and glutamine metabolism genes. Representative tracks display normalized H3K4me3 signal intensities across each gene locus.

(E) Western blot time-course analysis of H3K4me3 levels following DES treatment in control and shMUCP1 cells, showing progressive restoration of H3K4me3 in shMUCP1 cells.

(F) Quantitative RT-PCR analysis of MUC20-OT1 mRNA levels in shNC and shMUCP1 cells treated with DES over time, revealing a significant upregulation of MUC20-OT1 expression at 27 hours post-treatment.

(G) Western blot analysis of H3K4me3 levels in shMUCP1 cells cultured under glutamine-replete (+ Gln) and glutamine-deprived (− Gln) conditions, with or without DES and KDOAM-25 treatment.

(H) CRC cells were seeded in 6-well plates and treated with DMSO, diethyl succinate (DES), or KDOAM-25 for 24 hours. After 48 hours of total culture, supernatants were collected, and extracellular glutamine concentrations were measured using a glutamine assay kit. Glutamine uptake efficiency was calculated as: (Glutamine concentration in blank control medium − experimental group) / blank control. Data are presented as mean ± SD (n = 3). P-values were calculated using one-way ANOVA with multiple comparisons.

(I) Relative intracellular succinate levels in cells treated with DMSO, DES, or KDOAM-25 under glutamine-replete (+ Gln) and glutamine-deprived (− Gln) conditions for 24 hours. Intracellular succinate levels were measured using a succinate quantification kit and normalized to total protein content determined by BCA assay. Two-way ANOVA was used to assess the significance of differences in metabolite levels between groups.

Unexpectedly, H3K4me3 enrichment was also detected upstream of the MUC20-OT1 transcription start site (Fig. 4D). Notably, supplementation with exogenous succinate effectively restored the MUCP1-mediated decrease in H3K4me3 levels (Fig. 4E), prompting us to reconsider the regulatory interplay among MUC20-OT1, MUCP1, succinate, and H3K4me3. Strikingly, in MUCP1-silenced cells, MUC20-OT1 expression was significantly upregulated after diethyl succinate treatment for 27 h (Fig. 4F). These findings suggest that MUCP1 modulates cellular utilization of succinate and glutamine, while succinate, in turn, dynamically regulates the expression of both MUC20-OT1 and MUCP1.

Glutamine has been recognized as a primary contributor to succinate production in cancer metabolism [57,58]. To determine whether MUCP1-mediated succinate metabolism facilitates glutamine utilization, we treated MUCP1-deficient CRC cells with DES and KDOAM-25, a compound that enhances global H3K4me3 levels. Both agents markedly elevated H3K4me3 levels (Fig. 4G). Additionally, our data show that treatment with either DES or KDOAM-25 led to a marked increase in glutamine uptake by CRC cells (Fig. 4H). Notably, while KDOAM-25 elevated H3K4me3 levels under glutamine-deficient conditions, it was unable to restore intracellular succinate levels in the absence of glutamine. However, in glutamine-replete conditions, KDOAM-25 effectively facilitated the conversion of glutamine into succinate (Fig. 4I). These results suggest that MUCP1 contributes to maintaining the glutamine‑dependent metabolic state of CRC cells. This effect is likely mediated through succinate-induced H3K4me3 enrichment, which enhances the transcription of glutamine-metabolizing enzymes, thereby facilitating the conversion of glutamine into succinate and other nutrient intermediates necessary for tumor growth (Fig. S4H).

MUCP1 deficiency causes succinate accumulation in mitochondria

Tumor cells are highly dependent on glutamine and exhibit impaired growth under glutamine-deprived conditions [59,60]. Consistent with this, our results showed that MUCP1 knockdown reduced glutamine uptake (Fig. 5A), leading to a decrease in intracellular succinate levels (Fig. 5B). While our previous findings have suggested that MUCP1-mediated alterations in succinate influence glutamine assimilation, the precise molecular mechanism by which MUCP1 regulates succinate metabolism in cancer cells remains to be elucidated. We assessed succinate levels in mitochondrial and cytosolic compartments of CRC cells with transient MUCP1 knockdown. A marked accumulation of succinate within mitochondria was observed in ASO-treated cells, whereas re-expression of MUCP1-FLAG restored succinate redistribution to normal levels (Fig. 5C, D). Succinate, though primarily localized within the mitochondrial matrix, is known to escape into the cytoplasm and extracellular milieu during tumor progression, enhancing its oncogenic potential [25]. These observations led us to propose that MUCP1, residing on the outer mitochondrial membrane, may mediate the transport of excess mitochondrial succinate, thereby acting as a regulator of intracellular succinate dynamics in CRC cells.

Fig. 5.

The inhibition of MUCP1 results in impaired transport of succinate from mitochondria.

(A) Cells were cultured under standard conditions for 24 hours, after which culture supernatants were collected. Extracellular glutamine concentrations were measured using a glutamine assay kit, and glutamine uptake efficiency was calculated as: (Glutamine concentration in blank control medium − experimental group) / blank control. Data are presented as mean ± SD (n = 6). Statistical analysis was performed using unpaired two-tailed Student’s t-test.

(B) Cells were cultured for 24 hours in medium with or without glutamine, and an equal number of cells were collected to measure intracellular succinate levels. Two-way ANOVA was used to assess the significance of differences in metabolite levels between groups.

(C-D) HCT8 (C) and HCT15 (D) cells were transfected with antisense ASOs or the indicated plasmids (LV-NC or MUCP1-FLAG). Mitochondrial and cytosolic fractions were isolated 48 h after transfection. All procedures were performed rapidly and maintained at 4 °C to preserve metabolic stability. DEBM was added to the isolation buffers to minimize mitochondrial succinate efflux. Succinate concentrations were normalized to total protein content determined by BCA assay. Data are presented as mean ± SD (n = 6). Statistical analysis was performed using unpaired two-tailed Student’s T-test.

(E-G) Following ASO transfection, HCT8 and HCT15 cells were harvested at specified times for mitochondrial and residual component separation. Glutamine levels and protein concentrations were directly assessed in whole cells (E). Succinate levels were measured in mitochondrial (F) and extramitochondrial fractions (G) of CRC cells. Data were standardized to protein concentrations (n = 6). Statistical significance between groups was determined using one-way ANOVA. *p < 0.05, **p < 0.01, or ***p < 0.001, ns indicates no significance.

To verify whether MUCP1 governs mitochondrial succinate transport, we tracked succinate and glutamine dynamics over time in CRC cells following MUCP1 knockdown. Mitochondrial and cytosolic fractions were collected at 12, 24, 48, 72, 96, and 120 hours. A significant drop in intracellular glutamine levels emerged at 48 hours (Fig. 5E), coinciding with sustained mitochondrial succinate accumulation. Succinate levels in the mitochondrial fraction peaked at 48 hours in HCT 8 and at 72 hours in HCT 15 (Fig. 5F). Upon ASO degradation and subsequent re-expression of MUCP1, succinate was rapidly redistributed from the mitochondria to the cytosol, restoring the intracellular metabolic balance. The temporal coincidence between mitochondrial succinate accumulation and intracellular glutamine depletion suggests that impaired succinate transport triggers a compensatory increase in glutamine catabolism (Fig. 5E). In terms of cytosolic succinate, HCT 8 showed no significant changes, while HCT 15 exhibited a gradual decline. This difference may reflect altered succinate secretion dynamics under MUCP1 deficiency (Fig. 5G). Collectively, these findings demonstrate that MUCP1 plays a critical role in mitochondrial succinate transport and is essential for maintaining intracellular glutamine metabolic homeostasis. In line with our earlier results, ¹³C-labeled glutamine metabolic flux analysis revealed a substantial reduction in TCA cycle activity in shMUCP1 cells (Fig. 6A). MUCP1 loss led to decreased labeling of several key metabolites, including succinate (m+4), α-ketoglutarate (m+5), as well as fumarate (m+4) (Fig. 6B-D) and glutathione (m+5) (Fig. S5A). These observations indicate that MUCP1 deficiency disrupts mitochondrial succinate dynamics and compromises glutamine metabolism, suggesting a contributory role in maintaining tumor cell metabolism.

Fig. 6.

MUCP1 deficiency leads to metabolic reprogramming.

(A-D) Comparative relative abundance of ¹³C₅-glutamine-derived metabolites in the TCA cycle between MUCP1-deficient and control cells. Mass isotopolog distributions of succinate (B), α-KG (C) and Fumarate (D) are shown (mean ± SD, n=6).

(E) Histogram depicting the quantity of differentially expressed genes in SW480 cells with MUCP1-FLAG compared to control cells.

(F) Heatmap of sample correlations derived from transcriptome sequencing data.

(G) Schematic illustration of MUCP1-mediated metabolic remodeling in CRC cells.

As mentioned above, MUCP1 expression in SW480 cells showed no significant difference from that in normal colonic cells (Fig. 2E, F). We thus established a stable MUCP1-FLAG overexpression model in SW480 cells to assess transcriptional and metabolic consequences. Transcriptomic analysis revealed minimal changes, with only a few differentially expressed genes between MUCP1-FLAG and vector controls (Fig. 6E), and overall gene expression profiles remained highly correlated (Fig. 6F). Although KEGG enrichment pointed toward metabolic pathway involvement (Fig. S5B), MUCP1 overexpression did not promote colony formation (Fig. S5C) nor alter mitochondrial succinate levels (Fig. S5D). In SW480 cells, MUCP1 elevation resulted in negligible transcriptional or metabolic shifts and failed to confer an oncogenic growth advantage, indicating that MUCP1 is dispensable for tumor initiation and acts primarily as an adaptive factor under basal conditions. In line with previous findings, MUCP1 expression decreased in high-expressing CRC cells following metabolic stress induced by glutamine deprivation (Fig. S5E). These findings imply that while MUCP1 is insufficient to independently drive malignant transformation under steady-state conditions, its upregulation serves as a metabolic adaptation as tumors progress. Importantly, loss of MUCP1 compromises the succinate/glutamine metabolic balance, leading to metabolic reprogramming that may contribute to the suppression of tumor progression. (Fig. 6G).

MUCP1 exhibits sensitive responses to succinate levels and linked to SLC25A10 in modulating succinate homeostasis

It is well-established that SLC25A10, an integral membrane protein localized to the mitochondrial inner membrane, facilitates the transport of TCA cycle intermediates, including malate and succinate, from the mitochondria into the cytosol. Structurally, SLC25A10 functions as a homodimer, consisting of six transmembrane domains, with both the N- and C-termini exposed to the cytosol, while the remaining regions are embedded within the inner mitochondrial membrane [[61], [62], [63]]. Analysis of GEPIA database revealed that SLC25A10 expression is significantly upregulated in CRC tissues compared to normal tissues (Fig. S6A). Notably, a strong positive correlation between MUC20-OT1 and SLC25A10 expression was observed in normal colonic tissue (R = 0.41; Fig. S6B), while this correlation was markedly weakened following cancerous transformation (R = 0.12; Fig. S6C), suggesting that distinct regulatory mechanisms may govern their expression during tumorigenesis.

To clarify the regulatory relationship between MUCP1 and SLC25A10 in mitochondrial succinate transport, we performed co-immunoprecipitation (coIP) experiments to investigate their potential interaction. CoIP samples were separated by SDS-PAGE, and MUCP1-specific bands were visualized by silver staining and analyzed by mass spectrometry. Notably, several SLC family proteins were identified, and the interaction between MUCP1 and SLC25A10 was confirmed in CRC cells (Fig. 7A and B). Further analysis revealed that silencing SLC25A10 markedly reduced MUCP1 expression (Fig. 7C, D), while MUCP1 depletion also led to a significant downregulation of SLC25A10 (Fig. 7E). The results indicate that interfering with MUCP1 or SLC25A10 may compromise mitochondrial succinate transport, leading to insufficient cytosolic succinate signaling and subsequently altering the expression levels of both MUCP1 and SLC25A10. Notably, overexpression of MUCP1-FLAG in SLC25A10-deficient cells failed to rescue succinate efflux, indicating that functional SLC25A10 and its sufficient expression are indispensable for this process (Fig. 7F). Moreover, succinate transport defects induced by SLC25A10 knockdown led to reduced H3K4me3 histone modifications, supporting the notion that impaired succinate transport alters the epigenetic landscape (Fig. 7G). Given that SLC25A10 is a known mitochondrial succinate transporter, and its interaction with MUCP1 has been validated, our findings suggest that MUCP1 is functionally associated with SLC25A10 and may facilitate mitochondrial succinate transport through this coordination.

Fig. 7.

MUCP1 facilitates SLC25A10-mediated mitochondrial succinate transport.

(A) Co-immunoprecipitation was performed using anti-FLAG antibodies in CRC cells, followed by mass spectrometry to identify MUCP1-binding proteins. SLC family proteins identified in the analysis are highlighted in the results.

(B) SLC25A10 was confirmed as a MUCP1-interacting protein via western blot.

(C) Evaluation of SLC25A10 expression and knockdown efficiency following siRNA transfection in cells.

(D) SLC25A10 and MUCP1 expression levels were analyzed following siRNA-mediated knockdown of SLC25A10.

(E) Western blot analysis of SLC25A10 and MUCP1 expression in cells after knockdown of MUCP1.

(F) Mitochondrial and remaining cellular fractions were isolated under the indicated conditions, and succinate levels were measured and normalized to total protein content. Data are presented as mean ± SD (n=6). Statistical analysis was performed using unpaired two-tailed Student’s T-test.

(G) Western blot detection of H3K4me3 modifications under the indicated conditions to evaluate the impact of SLC25A10 and MUCP1 on histone methylation.

(H-K) The qRT-PCR analysis of relative expressions in MUC20-OT1 and SLC25A10 in HCT8 and HCT15 cells after treatment with DES (5 mM), DEBM (1 mM), and DMM (10 mM) for indicated periods.

(L-N) Western blot analysis of MUCP1 and SLC25A10 protein levels in HCT8 and HCT15 cells after treatment with DES (L), DEBM (M), and DMM (N).

To better understand the regulatory interplay between MUCP1 and SLC25A10 in mitochondrial succinate transport, we employed three distinct inhibitors to modulate intracellular succinate levels in CRC cells and examined the expression dynamics of MUCP1 and SLC25A10 (Fig. S6D). Treatment with DES, a membrane-permeable compound known to elevate cytosolic succinate levels [64], significantly increased the mRNA expression of both MUCP1 and SLC25A10 (Fig. 7H–K). This suggests that CRC cells rapidly respond to elevated cytoplasmic succinate by upregulating genes involved in succinate transport. Notably, MUCP1 protein showed a significantly higher than SLC25A10, indicating its more sensitivity to succinate (Fig. 7L). Upon treatment with diethyl butylmalonate (DEBM) [65], a mitochondrial succinate transporter inhibitor, MUCP1 and SLC25A10 proteins were gradually degraded over time, highlighting their coordinated function (Fig. 7M). Similarly, treatment with competitive SDH inhibitor dimethyl malonate (DMM) [23], which induces mitochondrial succinate accumulation, significantly upregulated MUCP1 expression without immediately affecting SLC25A10 protein levels (Fig.7N). Consistent with the effects observed with DES, MUCP1 exhibited a more rapid response to succinate fluctuations, as reflected by its prompt elevation. Upon extending the duration of DMM exposure, SLC25A10 expression increased in cells with intact MUCP1 but remained largely unchanged in MUCP1-deficient cells (Fig. S6E). This observation suggests that the adaptive response of SLC25A10-mediated transport may be functionally linked to MUCP1. While DMM, DES, and DEBM are valuable tools for modulating succinate dynamics, we recognize their potential for broader metabolic effects beyond specific transport inhibition. Within this context, our data suggest that MUCP1 appears to play a supportive role in the adaptive upregulation of SLC25A10 upon succinate accumulation, potentially facilitating succinate efflux in coordination with SLC25A10. This proposed regulatory axis may contribute to the adaptive metabolic response of CRC cells to altered succinate demand. Finally, we employed immunohistochemistry (IHC) to assess the in vivo expression levels of H3K4me3, SLC25A10, and other markers in xenograft mouse models injected with control or MUCP1 knockdown CRC cells. MUCP1 inhibition resulted in a marked decrease in H3K4me3, SLC25A10, and Ki67 expression compared to controls, accompanied by an increase in E-cadherin levels (Fig. S5F).

Together with our previous findings, these results support a model in which MUCP1 acts as a potential succinate-responsive factor that correlates with SLC25A10-mediated efflux and subsequent epigenetic modifications. Our findings suggest that this regulatory pathway may participate in the metabolic reprogramming of CRC by enhancing glutamine utilization to support tumor growth. Accordingly, MUCP1 is implicated as a potential target for both metabolic regulation and therapeutic intervention in CRC.

Discussion

Advances in ribosome sequencing and bioinformatics are revealing an increasing number of previously hidden small peptides and microproteins [66,67]. Recent studies have shown that LINC00467 encodes a mitochondrial membrane micropeptide that promotes CRC cell proliferation by enhancing ATP synthase assembly and increasing mitochondrial oxygen consumption, highlighting the crucial role of micropeptides in regulating cellular metabolism [18]. However, a significant portion of protein products derived from lncRNAs may still be overlooked, and how metabolic reprogramming regulates CRC progression remains unclear. In this study, by integrating ribosome profiling data, along with analyses of coding potential and functional predictions, we identified MUC20-OT1 as the only qualified candidate. Through a series of experiments, we confirmed that one transcript of MUC20-OT1 translates into a 144-amino-acid microprotein, which we named MUCP1. Strikingly, MUCP1 exhibits high sequence conservation and shares notable homology with SDHA, a canonical mitochondrial enzyme. Unlike SDHA, MUCP1 is localized to the mitochondrial outer membrane and plays a critical role in maintaining succinate homeostasis across mitochondrial boundaries, especially in metabolically active CRC cells. Thus, the discovery of MUCP1 provides new insights into the complex metabolic activities of CRC cells.

Tumor cells often rely on metabolic reprogramming to satisfy their excessive demands, leading to increased dependence on nutrients, especially glutamine, to which they develop an addiction [68]. In simple terms, glutamine acts as a carbon source for replenishing the TCA cycle, a nitrogen source for synthesizing nucleotides and other non-essential amino acids, and as the primary substrate for glutathione synthesis, which is vital in preventing oxidative stress in tumor cells [69]. Consistent with this, our metabolomics analysis of MUCP1-deficient cells showed not only a marked decrease in glutamine levels but also a reduction in essential amino acids like threonine and asparagine, along with key metabolites such as α-KG, succinate, and glutathione. Recent studies have shown that glutamine is essential for sustaining the proliferation of cancer cells with mitochondrial defects [70,71]. In addition to maintaining cell proliferation by inhibiting the production of mitochondrial-derived reactive oxygen species (ROS) via reductive carboxylation, glutamine also generates sufficient glutathione to prevent excessive ROS accumulation, as severe ROS stress can lead to cancer cell death [72]. Analysis of metabolites in the culture medium revealed that glutathione levels in the supernatant of shMUCP1 cells were significantly reduced compared to control cells, suggesting that intracellular glutathione synthesis may be impaired. This depletion implies that, in the absence of MUCP1, cancer cells may increasingly rely on extracellular glutathione uptake to mitigate oxidative stress and maintain redox balance. Abnormal succinate accumulation can drive the production of large amounts of ROS via reverse electron transport, which may partially explain the significant consumption of extracellular glutathione during MUCP1 deficiency to mitigate excessive cellular oxidation [73]. Furthermore, cancer cells consume substantial amounts of glutamine to significantly enhance lipogenesis, which promotes tumor growth, primarily because glutamine is essential for the activation of sterol regulatory element-binding proteins (SREBP) and the lipogenesis process [72,74]. We also observed that MUCP1 deficiency disrupts pathways associated with hormone secretion and lipid metabolism. These findings collectively indicate that MUCP1 deficiency leads to a breakdown of the glutamine metabolic network in CRC cells.

Extensive research has demonstrated that mitochondrial succinate accumulation can lead to its leakage into the cytoplasmic succinate pool, stabilization of HIF-1α, involvement in nuclear epigenetic modification, or extracellular secretion to mediate immune signaling, all contributing to cancer progression [23,25,58]. Consistently, we demonstrated that MUCP1 inhibition results in the accumulation of succinate in mitochondria, preventing its timely transport and leading to elevated upstream metabolites like α-KG and compensatory glutamine breakdown. Prolonged MUCP1 suppression ultimately led to a global reduction in intracellular succinate levels, resulting in metabolic insufficiency and diminished overall cellular activity. Based on these findings, we propose that MUCP1 is passively upregulated during CRC progression as an adaptive response to increased metabolic demands. Conversely, forced MUCP1 overexpression in metabolically stable CRC cells does not immediately confer a proliferative or tumor-promoting advantage.

Glutamine-derived α-KG and succinate play critical roles in signaling, inducing epigenetic reprogramming in cancer cells to regulate tumor progression [75]. In our study, we observed a significant increase in the α-KG /succinate ratio in shMUCP1 cells, prompting us to investigate whether succinate-mediated H3K4me3 modifications influenced the expression of glutamine metabolism enzymes, ultimately leading to the collapse of glutamine metabolism within the cell. Notably, glutamine is a major source of succinate in cancer cells, and disruption of glutamine metabolism further reduces succinate levels, exacerbating the decline in genomic H3K4me3 levels. However, the regulatory chain between glutamine, succinate, and MUCP1 is not perpetuated indefinitely. We speculate that, on the one hand, cancer cells will absorb and utilize a certain amount of glutamine based on their needs, ultimately achieving a degree of metabolic stability. On the other hand, the conversion of glutamine into succinate is limited, as it also needs to be converted into other metabolites required by the cell. Therefore, the regulatory relationship between glutamine, succinate, and MUCP1 will ultimately maintain metabolic homeostasis at a certain level, rather than a continuous positive feedback loop of promotion or inhibition.

Prior to the discovery of the new protein MUCP1, previous studies indicated that succinate in the mitochondria could be transported to the cytosol via the SLC25A10 carrier on the mitochondrial inner membrane (IMM), maintaining proper mitochondrial function [61,76]. The elucidation of MUCP1’s function further refines the regulatory mechanisms of succinate transport. In our study, we used three different inhibitors that modulate succinate levels, and both MUCP1 and SLC25A10 responded, although MUCP1’s response seemed more sensitive. Interestingly, SLC25A10 inhibition reduced mitochondrial GSH levels via mechanisms independent of its canonical transport function. Beyond GSH, SLC25A10 is critical for maintaining the homeostasis of various mitochondrial metabolites, including glutamate, isocitrate, pyruvate, and oxaloacetate [77]. The effects of MUCP1 inhibition closely resembled those of SLC25A10, indicating a strong functional correlation between them. The phenotypic effects of MUCP1 inhibition closely mirrored those of SLC25A10, suggesting a robust functional correlation between the two factors.

Conclusion

In summary, we identified MUCP1, a microprotein encoded by the lncRNA MUC20-OT1, as a auxiliary regulator of mitochondrial succinate transport. To meet the elevated metabolic demands of CRC cells, glutamine is avidly imported and catabolized into succinate and other oncometabolites. MUCP1 expression is upregulated during CRC progression and translocate to the outer mitochondrial membrane, where it facilitates the transport of accumulated succinate from the mitochondrial matrix. Extramitochondrial succinate, in turn, activates histone H3K4me3 modifications, which enhances the expression of glutamine metabolic enzymes, creating a positive feedback loop that sustains the high metabolic state of CRC cells. MUCP1 likely serves as a succinate-responsive sensor that fine-tunes SLC25A10-mediated succinate transport, coordinating metabolic adaptation with epigenetic remodeling. These findings highlight the potential of targeting metabolic regulators and tumor-derived metabolites as a promising strategy for CRC therapy. Moreover, the vast number of uncharacterized lncRNA-encoded microproteins presents an untapped source of mechanistic insights and clinically relevant targets in cancer biology.

Abbreviations

CRC	Colorectal cancer
lncRNA	Long non-coding RNA
ORF	Open reading frame
SDHA	Succinate dehydrogenase flavoprotein subunit A
TCA	Tricarboxylic acid cycle
DES	Diethyl succinate
DEBM	Diethyl butylmalonate
DMM	Dimethyl malonate
ASO	Antisense oligonucleotide
GSH	Glutathione
α-KG	Alpha-ketoglutarate
ROS	Reactive oxygen species

Funding

This work was supported by grants from the National Natural Science Foundation of China (Grant No. 82402717), Jiangsu Provincial Medical Key Discipline Cultivation Unit (JSDW202239), Nanjing Medical Key Laboratory of Laboratory Diagnostics, Natural Science Foundation of Jiangsu Province (BK20251730).

Availability of data and materials

The RNA-seq data in this study have been deposited at Gene Expression Omnibus database with accession number GEO: GSE281816. Any additional information required to reanalyze the data reported in this work paper is available from the lead contact upon request.

Declarations

Ethics approval and consent to participate

Matched fresh-frozen primary colorectal cancer tissues were obtained from patients at Nanjing First Hospital. Informed consent was obtained from each patient, and the acquisition of tissue specimens for our study was approved by the Institutional Review Board and Ethics Committee at Nanjing First Hospital (KY20220124-04).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Table S1. Detailed sequences of siRNAs, ASOs, shRNAs and sgRNAs utilized in transfection experiments.

Table S2. Primer sequences used for quantitative real-time PCR analysis of target genes in this study.

Table S3. Full amino acid sequences of MUCP1 encoded by two distinct MUC20-OT1 transcripts (ORF1 and ORF2). Both isoforms share an identical N-terminal region comprising the first 39 amino acids.

Figure S1. Comprehensive screening links MUC20-OT1 with tumor metabolism and coding potential.

(A) PhyloCSF scores for candidate lncRNAs.

(B) CPAT values for candidate lncRNAs.

(C) Multispecies amino acid sequence alignment of MUCP1 and SDHA visualized with the R package ggmsa.

(D) Functional prediction of MUC20-OT1 specific coding transcripts via the AnnoLnc2 database.

(E) High expression of MUC20-OT1 is significantly associated with shorter overall survival in CRC in the GEPIA cohort.

Figure S2. Antibody selection and localization studies for MUCP1.

(A) Diagram of homologous SDHA antibodies used for MUCP1. Blue lines indicate amino acid sequences at the N-terminus targeting MUCP1 precisely. Green lines indicate sequences targeting the middle region of MUCP1. Red text highlights amino acid sequences differing between MUCP1 and SDHA.

(B) Total protein was extracted from CRC cells, and gel slices corresponding to proteins below 20 kDa were excised. MUCP1-specific peptides were identified by Mass spectrometry.

(C) RNA-FISH assays of LncRNA MUC20-OT1 in HCT8 cells. Scale bars, 20 μm.

(D) Detection of MUCP1 expression in CRC cells using another homologous SDHA (Gly166) antibody.

(E) Mitochondrial particles extracted from HCT8 cells incubated with different concentrations of proteinase K solution followed by western blot analysis. Cox IV and TOMM20 serve as markers for the mitochondrial matrix and outer membrane, respectively.

(F) Mitochondria isolated from HCT8 cells were extracted with sodium carbonate. Western blot was performed to detect MUCP1, TOMM20, and HSPA9.

Figure S3. MUCP1 promotes CRC progression independently of its parent lncRNA

(A) Cell proliferation assessment in HTC8 and HCT15 cells post-transfection with ASOs or indicated plasmids via the CCK-8 method.

(B) Assessment of invasion and migration in HTC 8 and HCT15 cells via Transwell assay, conducted 24 hours post-transfection with ASOs or indicated plasmids (n = 3). Scale bars, 100 μm.

(C-D) RNA levels of MUC20-OT1 and protein levels of MUCP1 in HCT8 and HCT15 cells following shRNA-mediated knockdown via lentiviral infection.

(E) The in vivo growth of the indicated cell lines with stable MUCP1 knockdown was examined. Tumor volume over time in HCT8 and HCT15 xenografts, demonstrating reduced growth rates in MUCP1-deficient cells.

(F) Tumor weight in xenograft models of HCT8 and HCT15 cells with MUCP1 knockdown compared to control. Data are presented as the mean ± SD. Statistical significance was assessed using a two-sided unpaired Student's t-test.

Figure S4. The MUCP1-succinate axis modulates H3K4me3 enrichment at the promoters of glutamine metabolism-related genes.

(A) Succinate and α-KG levels in HCT8 cells from Figure 3B were used to calculate the α-KG/succinate ratio.

(B) H3K4me3 histone modification profiles at promoter regions of MUC20-OT1 and glutamine metabolism-related genes in CRC. Data obtained from the ENCODE database, project IDs ENCSR574USP and ENCSR577DVK.

(C-D) The relative mRNA levels of glutamine metabolism genes in HCT8 and HCT15 cells under the indicated conditions. Cells were treated with DES (5 mM), OICR-9429 (20 μM), or their combination for 24 hours.

(E-G) ChIP-qPCR analysis was performed to evaluate the fold enrichment of H3K4me3 at the promoter regions of GLS1 (E), GLUL (F), and MUC20-OT1 (G) in CRC cells.

(H) MUCP1 facilitates succinate transport, which enhances H3K4me3-driven expression of glutamine catabolic enzymes, forming a positive feedback loop that sustains glutamine metabolism in CRC cells.

Figure S5. Variations in MUCP1 result in metabolic changes in CRC.

(A) Relative abundance of ¹³C₅-glutamine-derived isotope-labeled glutathione between MUCP1-deficient and control cells.

(B) KEGG pathway analysis of differentially expressed genes between SW480 cells overexpressing MUCP1-FLAG and control cells.

(C) Colony formation assay of SW480 cells overexpressing MUCP1-FLAG compared to control cells (n = 3).

(D) SW480 cells (LV-NC and MUCP1-FLAG) were collected for the extraction of mitochondrial and residual components, followed by measurement of succinate levels and protein concentrations of each component. Data was normalized to protein concentrations.

(E) Western blot analysis of MUCP1 levels in HCT8 and HCT15 cells cultured in the presence (+) or absence (-) of glutamine.

(F) Immunohistochemical analysis of H3K4me3, Ki67, SLC25A10, and E-cadherin expression in shMUCP1 and control xenograft tumors. Scale bars, 100 μm.

Figure S6. Cooperative role of MUCP1 and SLC25A10 in mitochondrial succinate transport.

(A-C) GEPIA database analysis of SLC25A10 expression in colorectal cancer (A) and the expression correlation between SLC25A10 and normal colon tissue (B). and MUCP1 in CRC tissue (C).

(D) Schematic illustration of the mechanism of action of succinate-related inhibitors.

(E) Western blot analysis of SLC25A10 protein levels in the indicated cells with or without DMM treatment for 72 hours.

CRediT authorship contribution statement

Junjie Nie: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Data curation, Conceptualization. Xinwei Liu: Methodology, Investigation, Formal analysis. Mu Xu: Supervision, Resources, Conceptualization. Xinliang Gu: Methodology, Conceptualization. Shangshang Hu: Methodology, Data curation, Conceptualization. Huiling Sun: Validation, Supervision, Resources, Conceptualization. Linpeng Zhou: Resources, Formal analysis. Tao Xu: Supervision, Conceptualization. Yuqin Pan: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Conceptualization. Shukui Wang: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.