{"paper_id":"fb41c411-5609-40f6-a434-e9d7bf23e3a1","body_text":"communications medicine Article\nA Nature Portfolio journal\nhttps://doi.org/10.1038/s43856-025-01051-x\nA systems-based approach to uterine\nfibroids identifies differential splicing\nassociated with abnormal uterine\nbleeding\nCheck for updates\nChen-Yi Wang1,6,M a r t i nP h i l p o t t1,6, Darragh P O’Brien 2,A n n eN d u n g u3, Jessica Malzahn1,\nMarina Maritati1, Neelam Mehta1, Vicki Gamble1, Beatriz Martinez-Burgo 3, Sarah Bonham2,\nRoman Fischer 2,K u r t i sG a r b u t t3, Christian M. Becker 3,S a n j i vM a n e k3, Adrian L. Harris 4,\nFrank Sacher5, Maik Obendorf5,N i c o l eS c h m i d t5,J ö r gM ü l l e r5, Thomas M. Zollner5,\nKrina T. Zondervan 3,B e n e d i k tM .K e s s l e r2, Udo Oppermann 1 & Adam P. Cribbs 1\nAbstract\nBackground Uterine fibroids (UFs), benign tumours prevalent in up to 80% of women of\nreproductive age, are associated with signi ficant morbidity, including abnormal uterine\nbleeding, pain and infertility. Despite identification of key genomic alterations in MED12 and\nHMGA2, the pathogenic mechanisms underlying UFs and heavy menstrual bleeding (HMB)\nremain poorly understood.\nMethodsTo correlate systematically genetic, transcriptional and proteomic phenotypes, we\nconducted an integrative multi-omic approach utilising targeted DNA sequencing, RNA\nsequencing and proteomic methodologies, encompassing fibroid, myometrium, and\nendometrium tissues from 91 patients.\nResults In addition to con firming the presence of MED12 mutations, we identify variants in\nAHR and COL4A6. Multi-omic analysis of endometrium identi fies latent factors that\ncorrelate with HMB and fibroid presence with driver mutations of MED12, AHR, and\nCOL4A6, which are associated with pathways involved in angiogenesis, extracellular matrix\norganisation and RNA splicing. We propose a model, supported by in vivo evidence, where\naltered signalling of MED12-mutated fibroids influences RNA transcript isoform expression\nin endometrium, potentially leading to abnormal uterine bleeding.\nConclusions This study presents a comprehensive integrative approach, revealing that\ngenetic alterations in UF may in fluence endometrial function via signalling impacts on the\nRNA splicing mechanism. Our findings advance the understanding of complex molecular\npathways in UF pathogenesis and UF-associated endometrial dysfunction, offering insights\nfor targeted therapeutic development.\nHuman uterinefibroids (UF), also known as uterine leiomyoma, are benign\ntumours of the uterus that affect a l arge population of women of repro-\nductive age. They are particularly prevalent in black women in the United\nStates, with an incidence of approximately 80% for those aged between 35\nand 49, compared to 70% in white women of the same age group 1.U F s\ninterfere with normal uterine function, and in more than half of cases can\ncause distressing symptoms such as heavy menstrual bleeding (HMB),\npelvic pain, urinary incontinence, and/or infertility2. Despite the high pre-\nvalence of the condition, treatment options are hindered by the broad range\nof clinical manifestations. Symptomatic UFs are treated either by\nA full list of af filiations appears at the end of the paper. e-mail: Udo.oppermann@ndorms.ox.ac.uk; Adam.cribbs@ndorms.ox.ac.uk\nPlain language summary\nUterine fibroids are common benign non-\ncancerous tumours that grow in the womb\nand affect many women, often causing pain,\nheavy menstrual bleeding and problems with\nfertility. Genes are made up of DNA and are\ninherited. They provide instructions for mak-\ning proteins and RNA, other molecules within\nthe body. It is known that certain genes are\nassociated with people havingfibroids, but\nhow fibroids cause symptoms like heavy\nmenstrual bleeding is still unclear. We exam-\nined fibroid and endometrial tissues from 91\nwomen and looked at the DNA, protein and\nRNA present. We found changes infibroid\ntissues and discovered that these changes\nmay also affect nearby endometrial tissues,\nwhich line the womb. This can alter how\ngenes and proteins are expressed and may\nexplain why bleeding occurs. Thesefindings\nprovide new insight into how uterinefibroids\naffect the body and may help develop better\ntreatments to manage symptoms and\nimprove women’s health in the future.\nCommunications Medicine |           (2025) 5:318 1\n1234567890():,;\n1234567890():,;\n\ntherapeutic interventions due to U F-growth dependence on female sex-\nsteroid hormones, including selective progesterone receptor modulators\n(SPRM) such as Ulipristal acetate (UPA) and gonadotropin-releasing\nhormone (GnRH) agonist therapy\n3–5, or by surgery, including hystero-\nscopic/laparoscopic myomectomy, embolization, hysterectomy6.I nt h e\nUnited States alone, UFs are cited to be the cause of over 50% of\nhysterectomies\n7, and direct costs for their treatment have been estimated to\nbe between $4 –9 billion annually 8. Irregular heavy menstrual bleeding\n(HMB; or AUB, abnormal uterine bleeding) is the most common symptom,\naffecting up to 46% UF patients\n9.H M Bs i g n ificantly impacts quality of life as\na result of concurrent pain, anaemia, mood swings, and potential social\nembarrassment\n10–12. Despite its prevalence, the molecular mechanism\nlinking UFs to HMB remains poorly un derstood, limiting targeted treat-\nment options12.\nMutually exclusive driver mutations in the mediator subunit 12\n(MED12)13 and high-mobility group AT-hook 2 (HMGA2)14 genes occur\nin ~90% of UF cases. Med12 forms part of the Mediator Complex, which\nregulates transcription initiation and elongation by RNA polymerase II\n15,\nwhile Hmga2 protein binds to, and alters the structure of DNA, pro-\nmoting assembly of protein complexes that regulate transcription\n16.\nOther genetic contributors to UF include inactivation of fumarate\nhydratase (FH), a key enzyme of the Krebs cycle that promotes hypoxia\nwhen mutated\n17,18, and dysregulation of the aryl hydrocarbon receptor\n(AHR), which in fluences extracellular matrix (ECM) formation and\nTGF-β signalling19. Additionally, deletion of the collagen genes COL4A5\nand COL4A6 has been linked to familial UF cases 14,20. However, how\nthese mutations contribute to the development of UFs and associated\nsymptoms are not yet fully understood.\nSeveral studies have investigated the UF mechanism, primarily using\nmicroarrays to compare myometrium andfibroid, although the sample size\nof these early studies was limited21–28. Recent studies, such as Mehine et al.20\nfor example, analysed 60 UFs with different genetic drivers (e.g., MED12\nmutations, HMGA2 rearrangements, FH inactivation), revealing distinct\npathway alterations in Wnt, prolactin,and IGF-1 (insulin-like growth factor\n1) signalling. Proteomic approaches, despite small sample sizes of cohorts,\nhave highlighted roles for apoptosis, inflammation, and cytokine regulation\nin the development of UFs. Collectively, these studies suggest UF develop-\nment is linked to ECM, WNT- β-catenin and TGF- β3 signalling\npathways29–32.\nIn this study, we applied multi-omic approach of endometrium,\nmyometrium and fibroid tissues from 73 UF and 18 non-UF patients to\ninvestigate the molecular mechanism underlying UF pathology and asso-\nciated HMB. We identified key genomic alterations that provide insight into\nUF development. Integration of multi-omic factor analyses highlight the\ncontribution of ECM dynamics and RNA splicing to UF-associated endo-\nmetrial dysfunction. Differential t ranscript usage and single-cell tran-\nscriptomic profiling consistently point to aberrant TGF-β signalling and its\nrole in modulating alternative splicing in the UF-affected endometrium.\nOur study provides insights into the molecular mechanism underlying\nuterine fibroid (UF), particularly in relation to heavy menstrual\nbleeding (HMB).\nMethods\nPatient samples and tissue collection\nFibroid, myometrium, pseudocapsule and endometrium tissues were col-\nlected from 137 donors undergoing hysterectomy, myomectomy or\nTransCervical Resection of Fibroids (TCRF) at the John Radcliffe Hospital,\nOxford, in accordance with ENDOX study guidelines (09/H0604/58). All\nexperimental protocols were approved by the local Research Ethics Com-\nmittee (National Health Services (NHS) Research (NRES) Committee South\nCentral-Oxford). Informed written consent was provided by patients par-\nticipating in the study. In all cases, UF diagnosis was confirmed surgically\nand by histology. HMB status and use of hormone therapy was established\nfrom clinical notes and donor questionnaires. Menstrual cycle phase was\ndetermined by histopathology of the endometrium. Tissue samples were\ncollected immediately after surgery, snap frozen in liquid nitrogen, and\nstored at −80 °C. The majority of the fibroid samples analysed were from\nthe central region. However, pseudocapsule tissue was available in a limited\nnumber of patients, and where present, it was included in the study. Samples\ncollected by TCRF tended to be of poo r quality and yielded little or no\nendometrium, as did myomectomies, and surgeries performed by morcel-\nlation could not be reliably separated into individual tissue types. Overall,\ntissues from 91 donors were retained for this study and deemed suitable for\nthis study.\nSureSelect targeted sequencing\nDNA for SureSelect assays and SNP arrays was purified from fresh frozen\nsamples stored at −80 °C using a PureLink Genomic DNA Kit (Invi-\ntrogen) according to the manufacturer ’s instructions for mammalian\ntissue. Eluted DNA was quanti fied by NanoPhotometer (Implen) and\nstored at −20 °C until further use. Approximately 100 ng of each DNA\nsample was used to create Illumina sequencing libraries using a NEBNext\nUltra II FS DNA Library Prep Kit (New England Biolabs (NEB), E7805S).\nAfter PCR ampli fication with index primers, targeted DNAs were cap-\ntured and enriched by SureSelect XT HS Target Enrichment Kit ILM Hyb\nModule according to the manufacturer ’s instructions (Supplementary\nData 5, Agilent). Indexed libraries were quantitated by high-sensitivity\nDNA ScreenTape assay for TapeStation (Agilent), pooled at equimolar\nconcentration, and sequenced on a NextSeq 500 to an average of ~8\nmillion reads/sample. Reads were initially assessed for quality using\nFastQ Screen v0.14.0, FastQC v0.11.9 and MultiQC v1.5.dev0. Raw reads\nof each sample were mapped to hg38 using BWA v0.7.17 and merged\ninto a single bam file. For SNPs and small insertions/deletions (indels),\nvariant calling was performed using mpileup provided in bcftools v1.9\n33\nusing human genome GRCh38, with default Bayesian genotype\nlikelihood-based models and the parameters of minimum mapping\nquality as 20 and minimum base quality as 30, to detect variants. A\nlikelihood ratio test was used to infer the probability of a variant at each\nsite, and the QUAL score (phred-scaled p-value for the null hypothesis of\nno variant) was used to assess con fidence. Sites with QUAL ≥ 50 (cor-\nresponding to 99.999% con fidence) were considered as candidate var-\niants. Variant annotation, effect prediction and associated phenotypes\nwere performed by SnpEff\n34 and Ensembl Variant Effect Predictor 35.\nBulk RNA-sequencing\nTissue samples stored at −80 °C were cryomilled with Trizol without\nallowing the tissue to thaw. Briefly, one stainless steel end cap was inserted\ninto a polycarbonate cylinder and precooled in liquid nitrogen along with\nthe other cap and impactor. On dry ice, the impactor, 1.6 mL of Trizol, and\nthe tissue sample were added to the cylinder, which was capped and placed\nin the cryomill. The procedure was performed for 3 cycles of 2 min. Once\ncompleted, samples were transferred to a 50 mL centrifuge tube pre-chilled\non dry ice. When processing multiple samples, tubes were kept on dry ice or\nstored at−80 °C prior to downstream batch processing. Sample tubes were\nplaced in a 37 °C water bath until tha wed, vortex mixed, aliquoted into\n1.5 mL centrifuge tubes and stored at−80 °C if not proceeding immediately\nto RNA extraction. RNA extraction was performed using a Direct-zol RNA\nminiprep kit (Zymo Research) and on-column DNAse I digest, according to\nthe manufacturer’s instructions. Eluted RNA was quanti fied by Nano-\nPhotometer (Implen), quality checkedby high-sensitivity RNA ScreenTape\nassay for TapeStation (Agilent), and stored at−80 °C until further use. RIN\nvalues generally ranged between 3 to 5, typical of tissue samples, but sug-\ngesting some 3’ bias would be observed in the RNAseq.\nApproximately 100 ng of each RNA sample was used to create Illumina\nsequencing libraries using a NEBNext Ultra II Directional RNA Library\nPrep Kit for Illumina with NEBNext Poly(A) mRNA Magnetic Isolation\nModule (New England Biolabs) according to the manufacturer’s instruc-\ntions. Indexed libraries were quantitated by high sensitivity DNA Screen-\nTape assay for TapeStation (Agilent), pooled at equimolar concentration,\nand sequenced on a NextSeq 500 to an average of ~20 million reads/sample.\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 2\n\nAnalysis of bulk RNA-sequencing\nReads were initially assessed for quality using FastQ Screen v0.14.0, FastQC\nv0.11.9 and MulitQC v1.5.dev0. Raw reads of each sample were then merged\ninto a single file and pseudo-aligned to the human genome hg38 with\nKallisto 0.46.0. The samples with alignment rate lower than 60% were\nexcluded from downstream analysis. Using the count matrix produced by\nKallisto, differential expression analysis was performed by DESeq2 v1.35.0\n36\nfor comparisons with the clinical factors such as cycle phase, HMB,MED12\nstatus, and the technique factor like b atch effect. Functional analysis,\nincluding gene set enrichment analys is (GSEA) and over-representation\nanalysis (ORA) was done by R packages clusterPro filer 4.2.237,38.F o rd i f -\nferential transcript usage analysis,raw reads of the samples were pseudo-\naligned to gencode.v29.annotation.gtf by Kallisto, and the output abun-\ndance files were imported by tximport39 and then analysed by DRIMSeq40\nand stageR41. Genes with differential transcript usage that passed thefilter\n(p <0 . 0 5i nD R I M S e qa n dt h e n5 %o v e r a l lF D Ri ns t a g e R )w e r ei n c l u d e di n\nthe final output of DTU analysis. Downstream analysis, including sequence\nalignment, conserved domain search, predicted protein structure of enco-\nded protein isoforms, was performed using the tools msa\n42,r a g p43,a n d\nAlphaFold 344, respectively.\nUterine fibroid protein extraction\nF r o z e nU Fs a m p l e sw e r ec r y o m i l l e do nl i q u i dn i t r o g e ni n1 . 6m Lo fal y s i s\nbuffer comprising 6 M urea, 2 M Thiourea, 50% RIPA, 4% SDS, 100 mM\nDTT, and supplemented with proteas e and phosphatase inhibitors. To\nrelease protein bound to RNA and DNA, 1 μL of benzonase nuclease was\nadded to 500 µL of each thawed sample and incubated on ice for 20 min.\nDue to the inherent toughness of the UF tissue samples, each was subjected\nto three rounds of bead beating fo r2 m i na t4 ° Cf o rm a x i m u mt i s s u e\nd i s r u p t i o n .S a m p l e sw e r et h e ns p u nd o w nf o r5m i na t1 0 , 0 0 0g and 4 °C.\nThe supernatant was diluted (1:5) in water to achieve a final DTT con-\ncentration of 20 mM. Reduced samples were alkylated by adding IAA to a\nfinal concentration of 40 mM and incubated at room temperature for 1 hour\nin the dark. To remove SDS and other contaminants, all samples were\nsubjected to a protein extraction pr ocedure of alternating washes in\nmethanol, chloroform and water. To maximise protein recovery, pre-\ncipitated pellets were resuspended in 500 µL of 100 mM TEAB buffer,\nsonicated on ice for 5 min in a water bath, and vortexed at room temperature\nfor 30 min. The protein content of each UF sample was then determined\nusing a standard BCA assay.\nSample digestion, clean-up, and TMT-labelling\nSamples were digested in a 96-well format using the SMART Digest kit\nprovided by Thermo Fisher Scientific. Briefly, 150 µg of each lyophilized UF\nsample was resuspended in 50 µL of 100 mM TEAB and added to 150 µL of\nthe accompanying SMART Digest buffer. Frozen SMART Digest PCR strips\ncontaining immobilized trypsin beads were thawed and spun down at\n1000 g for 1 min, and at 4 °C. Samples (200 µL) were transferred into the\nappropriate PCR tube and incubated on a heated shaker for 180 min at 70 °C\nand 1400g. Upon completion, samples were spun down at 1000 g for 1 min.\nUF digests were cleaned-up with the aid of a vacuum manifold using the\nSOLAμ Solid-Phase Extraction (SPE) Plates provided with the kit. Samples\nwere loaded in a 1:1 ratio (v:v) with 0.1% TFA, followed by one wash with\n0.1% TFA. Peptides were eluted with 70% ACN into a 96-well collection\nplate and lyophilised to completion. For TMT-labelling, samples (~150 µg)\nwere resuspended in 100 µL of 100 mM TEAB. Approximately 10% of each\nsample was removed for the preparation of global pooled samples. For this,\ntwo concentrations were prepared to be included in each TMT 10plex\nlabelling reaction, one undiluted pool of all samples (1X Pool), and a five\ntimes diluted pool samples (5X Pool). Immediately before use, TMT label\nreagents were equilibrated to room temperature. To each 0.8 mg vial, 82μL\nof anhydrous acetonitrile was added and the reagent allowed to dissolve for\n5 min with occasional vortexing, before being gently centrifuged to gather\nthe solution. For each TMT labelling reaction, 41 μLo ft h eT M Tl a b e l\nreagent was added to each 100μL of UF sample. The reaction was allowed to\nproceed for 1 hour at room temperature before being quenched for 15 min\nwith 8 μL of a 5% hydroxylamine solution. For each TMT 10plex experi-\nment, an equivalent volume (140μL) of sample was combined, resulting in a\ntotal protein amount of approximately 1.5 mg in afinal volume of 1.4 mL.\nEach concatenated sample was desalted on a C18 solid-phase extraction\ncartridge (Sep-Pak Plus, Waters).\nHigh-pH reversed-phase pre-fractionation\nApproximately 1.5 mg of digested TMT-labelled material was subjected to\noff-line high-pH reversed-phase pre-fractionation using the loading pump\nof a Dionex Ultimate 3000 HPLC with an automated fraction collector and a\nXBridge BEH C18 XP column (3 × 150 mm, 2.5 μm pore size, Waters no.\n186006710). Peptides were separated over a 100 min gradient using two\nbasic pH reversed-phase buffers (A: ammonium hydroxide in 100% water,\npH 10; B: ammonium hydroxide in 90% ac e t o n i t r i l e ,p H1 0 ) .T h eg r a d i e n t\nconsisted of a 12 min wash with 1% B, then increasing to 35% B over 60 min,\nwith a further increase to 95% B over 8 min, followed by a 10 min wash at\n95% B and a 10 min re-equilibration at 1% B. Theflow rate was set to 200μL/\nmin, with fractions collected every 2 min throughout the run. In total, 50\nfractions were collected over the run, but samples were concatenated down\nto a final of 10 fractions by combining every 10th sample. Each fraction was\ndried down and resuspended in 30 μL of 2% ACN:0.1% formic acid for\nanalysis by LC–MS/MS.\nHigh performance Liquid Chromatography Tandem Mass\nSpectrometry (LC-MS/MS)\nLC-MS/MS analysis was performed using a Dionex Ultimate 3000 nano-\nultra high pressure reversed-phase chromatography system coupled on-line\nto a Q Exactive High Field (HF) mass spectrometer (Thermo Scienti fic).\nSamples were separated on an EASY-Spray PepMap RSLC C18 column\n(500 mm × 75 µm, 2 µm particle size; Thermo Scienti fic) over a 60 min\ngradient of 2–35% acetonitrile in 5% DMSO, 0.1% formic acid and at 250\nnL/min. The mass spectrometer was operated in data-dependent mode for\nautomated switching between MS and MS/MS acquisition. Full MS survey\nscans were acquired fromm/z 400–2000 at a resolution of 60,000 atm/z 200\nand the top 12 most abundant precursor ions were selected for HCD\nfragmentation. The resolution of MS2 fragment ion detection was also set to\n60,000. Fractions were loaded with adjusted sample volumes to analyze\n∼1 μgo nc o l u m n .\nProteomics Data Analysis\nMS raw data were searched against the UniProtKB human sequence data-\nbase (92,954 entries) and TMT 10plex quantitation performed using Pro-\nteome Discoverer software (v 2.3; Thermo Scienti fic). Search parameters\nwere set to include carbamidomethyl (C) as afixed modification, with TMT\n6plex, oxidation (M), and deamidation (NQ) set as variable modifications. A\nmaximum of 2 missed cleavages was allowed. TMT 10plex quantitation and\ndata analysis were performed in Perseus (v1.6.0.2), resulting in the gen-\neration of hierarchical clustering, principal component analysis, and Vol-\ncano plots. For PCA analysis, samples underwent log\n2 transformation and\nall missing values were removed. This was then followed by median sub-\ntraction normalisation. For the generation of volcano plots, an identical\nprocessing workflow was used, but only 50% of the missing values were\nremoved. The missing values that remained were imputed from the normal\ndistribution (width 0.3, down shift 1.8 ). Differentially regulated proteins\nbetween groups of interest were subject to gene ontology and pathway\nenrichment analysis using STRINGdb (https://string-db.org/). Shortlisted\ntargets were further assessed for their biological relevance and therapeutic\npotential in the treatment of UFs using TargetDB (https://pypi.org/project/\ntargetDB/).\nIntegration of transcriptomics and proteomics by Multi-Omics\nFactor Analysis (MOFA)\nIn addition of the metadata containing the clinical information related to\ndonors, the log-normalized count matrices of transcriptomics and\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 3\n\nproteomics (Supplementary Data 4) were used as the input data to\nMOFA45,46. In the proteomic data, features that contain more than 50%\nmissing values were removed. MOFA is anunsupervised statistical method\nto integrate multiple modalities of omics data and to identify latent factors\nthat capture sources of variation acr oss datasets obtained from different\nplatforms. The latent factors represent coordinated variation across data\nmodalities, but do not inherently have predefined biological meanings. A\nMOFA object was prepared using default settings and trained under a slow\nconvergence mode, with the number of factors suggested by the algorithm.\nThe likelihood for both the transcriptomic and proteomic data was both\ninferred as Gaussian.\nThe MOFA model was trained in a Bayesian framework, which\ndiffers fundamentally from classical regression models that rely on\np-values for inference. Instead of computing p-values to assess feature\nsignificance, MOFA applies sparsity-inducing priors and automatic\nrelevance determination (ARD) that allow the model to estimate the\nrelevance of each feature through posterior inference. In this context,\nfeature loading weights represent the strength and direction of con-\ntribution of each feature to a given factor. Using sparsity in the weights,\nloading weight of many features are exactly zero, indicating their irre-\nlevance to the factor, while only a subset of features has non-zero weights,\nmeaningfully contributing to latent factors. Thus, the selection of rele-\nvant features is not based on statistical signi ficance using classical\nregression, but on the magnitude of their contribution as inferred by the\nposterior distributions of the model.\nFor functional interpretation, gene set enrichment analysis (GSEA)\nwas conducted using built-in funct ion of MOFA with default setting,\nincluding “mean.diff” (difference in the average weight between fore- and\nback-ground genes) for gene set statist ic, a parametric t-test, Benjamini-\nHochberg procedure to adjust p-values factor-wise for multiple testing, and\nfalse discovery rate (FDR) threshold 0.1 for significant pathways. All features\nassociated with factors were used as input. Pathways enriched in both omic\nlayers were prioritised. Shared feat ures associated with factors in both\nmodalities, with absolute loading weights higher than a cut-off value of 0.3,\nwere visualised using by STRINGdb. While this threshold is not derived\nfrom p-value, it serves as an interpretable cutoff to highlight features with\nstronger associations. Overrepresented pathways were analysed via the\nEnrichr database\n47 using its R interface.\nKnown clinical, biological and technical covariates were correlated\nwith MOFA-inferred factors to support interpretability. These included\ngenotype information (e.g.,MED12 status and SNPs), fibroid occurrence,\ntissue type (e.g., UF or myometrium), menstrual cycle phase, HMB symp-\ntom, hormone treatment, and batch effects. While several latent factors\nshowed associations with these known variables, other may reflect unknown\nsources of variation for future investigation.\nNuclei preparation for single-cell RNAseq\nA petri dish, 50 ml centrifuge tube, scalpel and forceps were precooled on\ndry ice before pseudocapsule samples were removed from −80 °C and\nplaced in the petri dish. Typical sample sizes ranged from 100 –500 mg.\nTissue was cut into thin slices and tra nsferred to centrifuge tubes. If pro-\ncessing multiple samples, cut tissue could be stored at −80 °C until use.\nSample tubes were transferred to wet ice, 4 ml of ice-cold CST buffer\n(146 mM NaCl, 10 mM Tris-HCl pH 7.5, 1 mM CaCl\n2,1m MM g C l2,0 . 5 %\nCHAPS (w/v), 0.01% BSA (w/v), 4 μl/ml SUPERaseIN, 4 μl/ml RNasein\nPlus, 1 cOmplete protease inhibitor tablet (per 10 ml)) added and tubes\nplaced on a rotator for 10 minutes at 4 °C. Samples were passed through\n30 µm cell strainers (MACS SmartStrainer) into prechilled 15 ml collection\nt u b e so ni c e .S a m p l et u b e sw e r er i n s e dw i t h2m li c ec o l dP B S+ 1% BSA,\nwhich was added to the cell strainer. Cell strainers were rinsed with an\nadditional 2 ml ice-cold PBS+ 1% BSA and collection tubes centrifuged at\n500 g for 5 minutes at 4 °C. Supernatant was removed and pellet washed by\nresuspending in 10 ml ice cold PBS + 1% BSA, centrifugation at 500g for\n5 minutes at 4 °C, removal of supernatant and resuspension in 500 µl ice\ncold PBS + 1% BSA. A subsample of the nuclei preparation was incubated\nwith DAPI (1 µg/ml) for 5 minutes, added to a haemocytometer and\ncounted under a fluorescent microscope. Concentration of the nuclei was\nadjusted to ~1,000 cells/µl and used as input for analysis by 10X Chromium\nsingle cell gene expression.\nLibrary preparation and Sequencing of single-cell RNA\nsequencing\nChromium single cell gene expression (10X Genomics) was performed\nusing the Chromium Next GEM Single Cell 3’ GEM, Library & Gel Bead Kit\nv3.1, Chromium Next GEM Chip G Single Cell Kit and Single Index Kit T\nSet A according to the manufacturer ’s instructions, starting with 20,000\nnuclei as input. Resulting libraries were quantitated by TapeStation (Agi-\nlent), pooled at equimolar concentration and sequenced (Novogene (UK)\nLtd or Genewiz GmbH) on an Illumina NovaSeq 6000 using a S4 Reagent\nKit v1.5 to give ~30,000 reads/cell.\nAnalysis of Single cell RNA sequencing\nRaw sequencing data (fastqfiles) were processed using the scflow workflows\n(https://github.com/Acribbs/scflow). The Kallisto BUS/BUStools (v0.39.3)\nworkflow1 was implemented to pseudo-align the reads, with a K-mer size of\n31 base pairs. Homo sapiens (human) genome assembly GRCh38 (hg38)\nwas used to construct a reference tra nscriptome. Individual samples of\nsingle-nuclei or single cells were analyzed by the pipeline of quantnuclei or\nquantcells implemented in the scflow workflows, respectively. The output\nwas converted to single-cell experiment objects\n48 and then to Seurat objects\n(Seurat v4.0)49. Quality control and filtering were performed on the Seurat\nobjects; any cell with a mitochondrial ratio higher than 0.1, or fewer than 300\nfeatures was removed. Doublets in the samples were detected using the R\npackage scDblFinder\n50 and removed in the sc flow pipeline with Seurat\nclustering.\nTo integrate the endometrium samples with the published data, wefirst\nused the VST method provided by Seurat for variable gene selection and\napplied Harmony v1.04\n51 for batch correction. Highly variable genes that\naccount for cellular heterogeneity in each main cluster were used and cells\nwere aligned using Harmony. For cel l-cell communication, we applied\nCellChat (v1.4.0)\n52 with input of two matrices, log log-normalized count\nmatrix and a matrix of the cell label.\nTHESC decidualization\nThe cell line T HESCs was received from ATCC (ATCC ® CRL-4003TM)\ncertificated mycoplasma free. Cells were incubated in DMEDM/F-12 with\nbicarbonate and HEPES (Sigma Cat# D 2906) supplemented with 10%\nfoetal bovine serum (FBS, Charcoal stripped F6765-500ML), puromycin\n(500 ng/ml), and 1% ITS Premix Universal Culture Supplement (Corning\n354350). For the three-day experiment of decidualization, cells were seeded\nin 6-well plates for 40,000 cells per well and incubated overnight. At the next\nday (Day 0), the decidualization were induced by adding the following\nreagents into cell medium: Medroxyprogesterone 17-acetate (Sigma,\nM1629; final conc. 1.0 µM), E2 (estradiol,final conc. 10 nM; Sigma E1024),\n8-Br cAMP (8-Bromoadenosine 3 ′,5′-cyclic monophosphate, final conc.\n500 µM; Sigma B6386-100mg). In addition to stimulation for decidualiza-\ntion, cells were further treated with DMSO as mock, TGF- β (10 ng/ml;\nMillipore GF346) together with orwithout MEK inhibitor (BAY 1076672,\n100 ng/ml) since Day 0, depending on the experimental design. Cells were\nharvested on Day 3 using Direct-zol RNA MiniPrep kit (Cambridge\nBioscience, R2052).\nLibrary construction and sequencing of Nanopore long-read\nsequencing\n50 ng RNA of each sample were reverse transcribed and barcoded by using\nthe PCR-cDNA barcoding kit (SQ K-PCB111.24) and NEBNext Compa-\nnion Module (NEB E7180L). Libraries were then sequenced on the Nano-\npore PromethION platform.\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 4\n\nAnalysis of long read sequencing to identify transcript isoforms\nBase calling of fast5 files was done by Guppy ( https://github.com/\nasadprodhan/GPU-accelerated-guppy-basecalling) and converted to fastq\nformat. Fastq files were processed by the pipeline (pipeline_count) imple-\nmented in the work flow TallyTriN ( https://github.com/cribbslab/\nTallyTriN/tree/main) and raw count matrix for PCA analysis was then\ngenerated. Reads of each sample were then aligned to hg38 genome by\nMinimap2\n53 w i t h--M Dflag enabled and output as SAM format. The SAM\nfile of each sample was processed by TALON54 v5.0 and Swan55 v2.0, using\ndefault settings, and gencode.v29.annotation.gtf as reference for isoform-\nlevel analysis (transcript switching genes and transcripts which are not in the\ngencode database due to alternative splicing) and visualisation.\nMice\nFemale mice (Balb/c) (purpose-bred animals, Janvier Labs) aged ~9 weeks\nwere housed according to the EU guideline 2010/63 EU. The study (study\ncode: A0384/09) was approved by the German animal welfare authorities\n(LAGeSo, Berlin).\nMouse model of menstruation and treatment regimens\nThe experimental model of menstr uation in mice was adapted from\nestablished protocols56. Female Balb/c mice were acclimatized to the animal\nfacility for one week before being trained in animal handling for at least one\nweek prior to inclusion in the study. Cage enrichment, such as nesting\nmaterial and hiding structures, wa s provided throughout the study to\nimprove the well-being of the animals. Mice underwent bilateral ovar-\niectomy, with analgesia provided before and after surgery by administering\ntramadol (1 mg/ml) orally via drinking water. Overall, the study is regarded\nas mildly burdensome, with no standard need for additional analgesic\ntreatment.\nOne week post-surgery, mice received subcutaneous injections of\n100 ng 17α-estradiol (E2) dissolved in a 1:9 ethanol to peanut oil solution\nfor three consecutive days. After a three-day interval, a subcutaneous\nsilastic implant delivering progesterone (P4, internal source\n57; 0.5 mg P4/\nday) was inserted dorsally. Concurrently, 5 ng of E2 was administered\ndaily for three consecutive days. On thefinal day of E2 treatment, 50 μlo f\nsesame oil was injected into one uterine horn to induce decidualization.\nFour days later, the P4 implant was removed to trigger progesterone\nwithdrawal.\nTo assess menstrual-like bleeding, tampon-like cotton pads (4–4.8 mm\nin diameter) were inserted into the vagina of mice at the time of P4 with-\ndrawal. Mice were fitted with paper collars to prevent the removal of the\npads. Tampons were replaced twice daily, and samples from each mouse\nwere collected individually. Blood volume was quantified using the alkaline\nhematine method\n58.B r i efly, tampons were first dried at room temperature\nand then immersed in 1000 ml of 5% sodium hydroxide (NaOH, w/v)\novernight under rotation at room tempe r a t u r et od i s s o l v eh a e mc h r o m o g e n .\nThe optical density of the haem-containing eluates was measured at 546 nm\nusing an ELISA plate reader. Blood volume contained in cotton swabs was\nmeasured based on a standard curve prepared from venous blood.\nSeventy-two hours after P4 withdrawal, mice were euthanized, under\ndeep terminal anaesthesia with iso flurane (>3%), by terminal blood col-\nlection from the vena cava. Uterine ti ssues were collected, weighed, and\nprocessed for further analyses. All surgical interventions were conducted\nunder isoflurane-induced anaesthesia, with pain prevention provided by\ntramadol treatment. Notably, no animals in this study experienced unex-\npected severe events or required rescue analgesic treatment, and no animals\nwere excluded from the experiment orfinal analysis. Mice were randomly\nallocated to placebo and treatment groups, and the treatment of the animals\nwas not blinded, as the primary readout, the quantitative ex vivo mea-\nsurement of blood loss, was performed blinded to the operator.\nTreatment in the mouse model\nGroups (n = 10) were treated with either the MEK inhibitor (BAY MEKi,\ncpd 2659, Bayer AG, Germany) at doses of 0.5 mg/kg/d p.o. or with the\nACVR1 inhibitor (TP-0184, Toledo Pharmaceuticals, USA) at doses of\n15 mg/kg/d p.o. dissolved in N-methyl-2-pyrrolidone (NMP)/ polyethylene\nglycol 400 (PEG400) (1/9) (d0-d15) in a volume of 5 ml/kg. Controls were\ntreated with vehicle alone qd/p.o.\nStatistics and reproducibility\nTranscriptomics and proteomics datamatrices used as input for the multi-\nomics factor analysis (MOFA) are provided in the Supplementary Data 4.\nPrior to analysis, proteomics data wasfiltered to retain features detected in at\nleast 50% of samples and then normalised and log transformed. Tran-\nscriptomics data were normalized and variance-stabilising transformed\nusing DESeq2. The MOFA model was trained with default parameters,\nincluding Gaussian likelihoods, sparsity priors like spikeslab_weights and\nard_weights, and a slow convergence setting (corresponding to an ELBO\ntolerance of 5e-8). The number of latent factors was inferred based on\nMOFA model performance.\nClinical information (Supplementary Data 3) and genotype informa-\ntion (Supplementary Fig. 2) were used to investigate the biological relevance\nof each factor. Unlike classical regression models that rely on frequentist\nstatistical significance (e.g., p-values), MOFA operates within a Bayesian\nframework that estimates the relevance of each feature using sparsity-\ninducing priors. Most features have zero contribution (loading weight is\nzero), while a subset of features with non-zero loading weights meaningfully\ncontribute to the latent factors. As a result, MOFA does not calculate\np-values for feature-fac tor associations. Instead, the magnitude of the\nloading weight of each feature on a latent factor indicates its importance and\ndirection of contribution. While some Bayesian models report posterior\ninclusion probabilities to quantify confidence in feature inclusion, MOFA\nidentifies relevant features based on their inferred weights. In this study,\nfeatures with absolute loading weight higher than a cut-off value of 0.3 in\nboth modalities were considered highly biologically relevant and selected for\nvisualisation in STRINGdb.\nFor Fig. 5c, two independent in vivo experiments were conducted to\ninvestigate the effects of TP-0184 (an ACVR inhibitor) and BAY-533 (a\nMEK inhibitor). Data analysis (Supplementary Data 6) was performed\nusing GraphPad Prism 10 software. For comparisons between the two\nrespective groups, statistical signi ficance was assessed using a one-sided\nStudent’s t-test ( p < 0.05; ****p < 0.0001). Based on extensive prior\nexperience with this model, the data particularly regarding bleeding as\nthe primary endpoint are considered robust. Due to ethical constraints\nand in agreement with the established reliability of the model, repetition of\nthe animal experiments was not approved by the local regulatory\nauthorities.\nResults\nClinical features of the cohort\nA total of 91 patients, predominantly European population, undergoing\nhysterectomy, myomectomy or trans-cervical resection of fibroids\n(TCRF) were recruited (Supplementary Fig. 1; Supplementary Data 1).\nThe majority had uterinefibroids (UFs), while 18 non-UF patients served\nas a comparative cohort, though they were not considered as healthy\ncontrols. These patients underwent surgery for conditions including\nendometriosis, adenomyosis, ovarian cysts or cervical neoplasia. Heavy\nmenstrual bleeding (HMB) status was determined via patient ques-\ntionnaires and clinical records, with 33 donors classi fied as HMB (Sup-\nplementary Fig. 1). As hormone treatment in fluences HMB symptoms,\npatients undergoing such treatment at the time of surgery were assumed\nto have therapeutic intervention. Menstrual cycle phase was primarily\ndetermined histologically, with clinical notes and hormone levels used\nwhen histology was unavailable. Notably, 35 patients had inactive\nendometrium due to hormone treatment. The collected tissues encom-\npassed distinct uterine compartments, including endometrium, myo-\nmetrium, fibroid, as well as pseudocapsule, a vasculature-rich region that\nsurrounds the tumour, which is not formed in all fibroids (Supplemen-\ntary Data 1).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 5\n\nGenomic insights into UF pathology\nTo investigate UF-associated genetic alterations, we performed targeted\nsequencing of key UF driver genes, including MED12, HMGA2, FH,\nCOL4A5/6, HMGA160, RAD51B14, AHR61, CAPRIN161, CUX162, DCN61 and\nPCOLCE63 candidate genes. The variant calling analysis focused on single\nnucleotide polymorphism (SNPs) and short indels. A likelihood ratio test\nwas used to infer the probability of a variant at each site, and sites with\nQUAL score (phred-scaled p-value forthe null hypothesis of no variant)≥\n50 (corresponding to 99.999% confidence) were considered as candidate\nvariants (Supplementary Data 2). Among 73 fibroids, 39.7% harboured\nMED12 variants, which are canonical UF mutations in intron 1 and exon 2,\nwith other MED12 variants having minimal fun ctional impact (Supple-\nmentary Fig. 2a). Furthermore, we identified mutation hotspots inCOL4A6,\nAHR and CUX1, including in-frame insertion-deletion and frameshift\nmutations inCOL4A6 exon 24, and missense variants inAHR exon 10 and\nCUX1 exon 16 (Supplementary Fig. 2b, upper, middle and bottom panel,\nrespectively).\nDifferential gene expression in UF HMB endometrium\nTo investigate gene expression profiles in the endometrium of UF patients\nwith HMB, we applied bulk RNA sequen cing and performed differential\ngene expression analysis using DESeq2. Principal component analysis\n(PCA) (Fig.1a) exhibited distinct separation between HMB and non-HMB\npatients with active menstrual cycle, along the PC1 and PC2 axes. Gene set\nenrichment analysis (GSEA), using a p-value cutoff of 0.05 and\nBenjamini–Hochberg (BH) adjustment for multiple testing, revealed that\nduring the proliferative phase, the gene expression profile was dominated by\ncell cycle and mitotic processes (Fig. 1b, left panel), whereas during the\nsecretory phase, immune-related pathways including in flammatory\nresponse and allograft rejection, as well as RAS signalling, were enriched\n(Fig. 1b, right panel; Fig.1c). These findings are consistent with established\nroles of inflammatory processes and leucocyte trafficking in the endometrial\nphysiology\n64. IL11 and LIF for example, were significantly upregulated in\nHMB patients (log2 fold change (log2FC) > 1.0, adjusted p-value (padj) <\n0.05), particularly in the secretory phase (Fig.1c, d). Recombinant human\n−10\n0\n10\n20\n−20 0 20\nPC1: 49% variance\nHMB status\nHMB\nNo HMB\nPC2: 16% variance\n01 0 2 0 3 0 4 0\nCount\nCount\nProliferativea b\nd\nc\nGL YCOL YSIS\nESTROGEN\nRESPONSE LATE\nSPERMATOGENESIS\nMITOTIC SPINDLE\nE2F TARGETS\nG2M CHECKPOINT\n0.04\n0.03\n0.02\n0.01\np.adjust\n40\nSecretory\n01 0 2 0 3 0\nESTROGEN\nRESPONSE EARL Y\nIL2 STAT5 SIGNALING\nCOAGULATION\nESTROGEN\nRESPONSE LATE\nALLOGRAFT REJECTION\nTNFA SIGNALING VIA NFKB\nKRAS SIGNALING UP\nINFLAMMATORY RESPONSE\n0.006\n0.004\n0.002\np.adjust\nINFLAMMATORY\nRESPONSE\nKRAS\nSIGNALING\nUP\nTNFA\nSIGNALING\nVIA NFKBALLOGRAFT\nREJECTION\nESTROGEN\nRESPONSE\nLATE\nAQP9\nCCL20\nIL2RB\nRASGRP1\nINHBA\nOSM\nNPFFR2\nCXCL10\nF3F3F3F3F3F3F3F3F3F3F3F3F3F3F3F3F3\nGPR132\nHPN\nCCR7\nPDE4B\nRGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1RGS1\nLIF\nOLR1\nIL1R1\nLAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3LAMP3\nSELLSLC1A2\nTNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10TNFSF10\nIL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18IL18\nIL7R\nIL10\nGPR183\nTNFSF15\nFPR1\nSLC28A2\nSEMA4D\nPLAUR\nMSR1\nNOD2\nCMKLR1\nADORA2B\nMYC\nCCL2\nBTG2\nTACR1\nCDKN1A\nSELE\nMMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10MMP10\nSLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPISLPI\nWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7AWNT7A\nMMP11\nTSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1TSPAN1\nPIGR\nHKDC1\nSOX9\nGALNT3\nF2RL1\nMMP9\nPLEK2\nFGF9\nG0S2\nIRF8\nCXCR4\nIL1RL2\nIL2RG\nPRDM1\nSCG5\nLCP1\nCFB\nIKZF1\nIGF2\nFCER1G\nMAP7\nKLF4\nPRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2PRKG2\nTSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7TSPAN7\nPLVAP\nCPECPECPECPECPECPECPECPECPECPECPECPECPECPECPECPECPE\nSNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25SNAP25\nPCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4PCP4\nLAMB3\nFJX1\nDUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4DUSP4BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1BCL2A1\nIER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3IER3\nSLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6SLC16A6\nTNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2TNFAIP2\nAREG\nDUSP1\nSLC2A3\nNR4A1\nZFP36\nJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNBJUNB\nIER5\nKLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2KLF2\nFOS\nSOCS3\nJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUNJUN\nATF3\nEGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1EGR1\nGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMBGZMB\nGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMAGZMA\nTRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1TRAT1 CD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79ACD79A\nCD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7CD7\nIL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11IL11\nPRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1PRF1\nCD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247CD247\nPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRCPTPRC\nCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LGCD40LG\nIRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4IRF4\nCXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3CXCR3\nLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTBLTB\nIL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7IL7\nNCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1NCR1\nCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAMCRTAM\nCD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86CD86\nHLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1HLA−DQA1\nCD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2CD2\nIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RAIL2RA\nCCR1\nCD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3ECD3E\nCD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96CD96\nITKITKITKITKITKITKITKITKITKITKITKITKITKITKITKITKITK\nZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70ZAP70\nF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2RF2R\nSTAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4STAT4\nPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCBPRKCB\nFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNAFLNA\nAGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2AGR2\nSERPINA1\nGALGALGALGALGALGALGALGALGALGALGALGALGALGALGALGALGAL\nOVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2OVOL2\nSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFNSFN\nST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14ST14\nGJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3GJB3\nSERPINA5\nST6GALNAC2\nCYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11CYP4F11\nCDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1CDH1\nTJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3TJP3\nSCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1ASCNN1A\nPKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3PKP3\nCXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14CXCL14\nLAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2LAMC2\nTFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2TFPI2\nKIF20A\nSLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5SLC7A5\nTMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3TMPRSS3\nMYB\nTOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2ATOP2A\nTFAP2C\nOPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3OPN3\nCCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1CCNA1\nSLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1SLC29A1\nTPSAB1\nHSPB8\nPTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3PTGER3\n−2.5\n0.0\n2.5\n5.0\n7.5\nlog2 (Fold Change)\n0.5\n1.0\n1.5\n2.0\n2.5\nHMB No\nProliferative\nTranscripts Per Million (TPM) 0\n2.5\n5.0\n7.5\n10.0\nSecretory\nHMB\nHMB No\nNo\nTranscripts Per Million (TPM)0\n10\n20\n30\n5\n10\nHMB NoHMB No\nProliferative Secretory\nLIF expression\nHMB No\nIL11 expression\nFig. 1 | Differentially expressed genes and enriched pathways in active endo-\nmetrium from UF patients. a Principal component analysis (PCA) plot showing\nclustering of endometrium samples from UF patients with or without heavy men-\nstrual bleeding (HMB) symptoms ( n = 15; proliferative phase: 4 HMB and 4 non-\nHMB; secretory phase: 4 HMB and 3 non-HMB).b Bar plots of pathways enriched in\nHMB endometrium, identi fied by gene set enrichment analysis (GSEA). Pathways\nare ranked by adjusted p-value, shown in a gradient of blue to red. Left and right\npanels correspond to the proliferative and secretory phase, respectively. c Network\nvisualisation of differentially expressed genes (DEGs; absolute log 2FC ≥ 1.5, padj <\n0.05) in secretory HMB endometrium, depicting the linkages of gene functions and\npathway associations. d Boxplots of IL11 and LIF expression in the proliferative (left\npanel) and secretory (right) phases of the endometrium.\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 6\n\nIL11 has shown > 50% reduction65 (ClinicalTrials.gov ID: NCT00524342) in\npictorial blood assessment chart (PBAC)66, which is widely used to assess\nmenstrual blood loss, implying a role for immune dysregulation on HMB.\nMulti-omic factor analyses identify dysregulation of ECM and\nRNA processing as key contributors to UF-associated HMB\nsymptoms\nTo gain deeper insight into UF-associated HMB, we applied multi-omic\nfactor analysis (MOFA)45,46, an unsupervised method that integrate bulk\ntranscriptomics, proteomics and genomics, and identifies latent factors that\ncapture sources of variation across datasets obtained from different plat-\nforms. We examined whether the identified latent factors were associated\nwith the known clinical, biological andtechnical variables including patient\nt y p e( U Fo rn o n - U Fp a t i e n t ) ,H M Bstatus (with or without HMB symp-\ntom), mutations identified in fibroids, and sample batches. MOFA analysis\nof 31 endometrial samples (UF and non-UF patients; Supplementary\nData 3-4) identified 7 latent factors (Fig. 2a, c). Factor 1 was signi ficantly\ncorrelated with HMB and hormone treatment (padj < 0.01), suggesting a\nlasting impact of therapeutic interventions on endometrium function.\nFactor 2 was strongly associated withfibroid presence (padj < 0.001; Sup-\nplementary Fig. 3), indicating the influence offibroid tissue on physiological\nfunctions of endometrium. Factor 7 was correlated with not only the pre-\nsence but also genomic alterations offibroid, includingMED12 UF muta-\ntions, AHR rs2066853 and COL4A6 rs6622312, all of which were also\ncorrelated with HMB (padj < 0.05; Fig.2ar i g h t&F i g .2b, c).\nThe relevance of each feature to a latent factor is identified by MOFA\nvia a Bayesian framework with sparsity-inducing priors. The contribution of\neach feature is inferred as a loading weight by posterior distribution of the\nMOFA model\n45,46. Loading weight of irrelevant features is exactly zero, while\nfeatures with non-zero loading weights on a latent factor indicate the strength\nand direction of contribution. GSEA analysis of Factor 1-associated features\nrevealed the enrichment in coagulation, angiogenesis and ECM organisation\n(false discovery rate (FDR) < 0.1) in both omics (Fig. 2d). Most of these\nfeatures with stronger association with Factor 1 (absolute loading weight≥\n0.3) were negatively association with HMB (Fig. 2e). For example CD59,\nwhose genetic deficiency is linked to haemolytic anaemia and thrombosis\n67,\nand angiogenin (ANG), an RNAase A superfamily member involved in\nneovascularization68, were downregulated in HMB endometrium.\nPathway analysis of Factor 2 and Factor 7 identi fied enrichment of\nRNA processing and metabolic process, including mRNA splicing and RNA\n3’-end processing (FDR < 0.1; Fig. 2f, g; Supplementary Fig. 3). These\nfindings suggest UF-induced dysfunction of RNA homoeostasis and the\nsubsequent aberrant splicing eventsin endometrium may be exacerbated by\nMED12, AHR or COL4A6 variants in fibroids, potentially contributing to\nHMB symptom.\nIntegrated analysis of fibroid ( n = 50) and myometrium ( n = 41,\nincluding 31 UF and 10 non-UF patients) identified 6 latent factors (Sup-\nplementary Fig. 4). Factor 2, unaffected by batch effects, strongly correlated\nwith tissue type (padj < 0.001; Supplementary Fig. 4a–c) and was associated\nwith pathways related to ECM and collagenfibril organisation, angiotensin\nmaturation, and hormone metabolic process (FDR < 0.1; Supplementary\nFig. 4d, e). UCHL1, a ubiquitin C-terminal hydrolase involved in protein\nhomoeostasis, was positively associated withfibroid tissue (Supplementary\nFig. 4e) and has been implicated in promoting TGF-β signalling via stabi-\nl i s a t i o no ft h et y p eIT G F -β receptor\n69. Higher level ofUCHL1 in UFs70 may\ncontribute to the elevated TGF- β signalling. These findings were further\nsupported by 2D Annotation Enrichment analysis71 (Supplementary Fig. 5),\nreinforcing the central role of ECM dysregulation in fibroid pathology.\nMOFA analysis of fibroid samples showed a noteworthy albeit weak cor-\nrelation with MED12 UF mutations, AHR rs2066853 and COL4A6\nrs6622312 at Factor 5 (padj < 0.01; Supplementary Fig. 6a –c). Enriched\npathways by GSEA also highlighted ECM, collagenfibril organisation and\nangiogenesis, addressing the crucial role of ECM in UF pathology (FDR <\n0.1), and indicating these variant s may exacerbate ECM dysregulation\n(Supplementary Fig. 6d, e).\nDifferential transcript usage reveals the role of TGF-β signalling\nand RNA processing in UF endometrium pathology\nOur integrated analysis identified RNA processing and mRNA splicing as\nkey molecular mechanisms underlying UF endometrium pathology. To\nexamine transcript-level alterations in the endometrium of UF patients with\nHMB, we performed differential transcript usage (DTU) analysis on active\nendometrium samples, excluding those under therapeutic hormone treat-\nment to minimise confounding effec ts. Using DRIMSeq for initial DTU\ndetection (p-value < 0.05) and stageR for further statistical testing (overall\nfalse discovery rate (OFDR) < 0.05)), we identified 684 transcripts across\n478 genes in differential transcript usage between HMB (n = 8) and non-\nHMB (n = 7) patients. Alternative transcript usage was observed in genes\nincludingTGFBR2, ENG, NRP1, TBXA2R,a n dPDE1A,w h i c ha r ei n v o l v e d\nin blood vessel morphogenesis and ang iogenesis, prostaglandin synthesis\nand regulation, and calmodulin-mediated signalling, respectively (Fig.3a).\nPathway enrichment analysis highli ghted processes related to vascular\nsmooth muscle cell differentiation, peptide antigen assembly with MHC\ncomplexes, and ribosome biogenesis (Fig.3b).\nWhen comparing endometrial samples from patients with MED12-\nmutant fibroids (n = 10) and those with wild-type MED12 (n = 5), 2,784\ntranscripts across 2134 genes ( p- v a l u e<0 . 0 5 a n d O F D R<0 . 0 5 ) w e r e\nidentified. We observed DTU in genes involved in protein modi fication,\nstress-activated MAPK cascade, mRNA transport and RNA splicing such as\nHNRNPRand HNRNPL(Fig.3c, d). Notably, TGF-β signalling emerged as a\nkey pathway, with DTU analysis identifying alternative usage inTGFBR1,\nTGFBR2 and TGFBR3 in TGF-β receptor signalling pathway, as well as\nANGPT1 and ANGPT2 in angiogenesis (Fig. 3ca n de ) .T h e s efindings\nunderscore the roles of TGF- β signalling in UF-induced dysfunctions in\nendometrium, particularly in the presence ofMED12mutations. Given that\nSMAD3-mediated TGF-β signalling directly regulates alternative splicing\n72\n,73, the observed DTU of TGF- β receptors may impact downstream sig-\nnalling dynamics. A striking example is TGFBR2, which encodes two\nalternative spliced variants, T βR-II and T βRII-B, with distinct\nligand-binding affinities. T βR-II, which binds TGF- β1/3, and T βRII-B,\nwhich binds TGF- β272,73. Intriguingly, our analysis found that T βRII-B\n(ENST00000359013) was the dominant isoform in the endometrium of\npatients with HMB or with MED12-mutant fibroids (Fig. 3a and e). As\ns h o w ni nS u p p l e m e n t a r yF i g .7 ,a na d d i t i o n a lp e p t i d ec o m p o s e do f2 5\namino acid residues in the extracellular domain of TGF-βRII\n74 alters TGF-\nβRII protein structure, suggesting a shift in TGF-β ligand specificity in this\npathological context.\nMost genes identified through DTU analysis did not exhibit differential\nexpression at the gene level (padj < 0.05, absolute log2FC ≥ 1.5), with only a\nsmall subset overlapping between D TU and differential gene expression\nanalysis. In addition to angiogenesis, DTU analysis identified genes involved\nin prostaglandin synthesis (PTGES, PTGES2,a n d PTGFR), progesterone\nsignalling (PGR), and FGF signalling ( FGF7 and FGFR2). These findings\nfurther highlight transcript-level regulation as a crucial layer of molecular\ncontrol in UF pathology and suggest thatalternative splicing may contribute\nto UF-associated symptoms like HMB.\nSingle-cell transcriptomic analysis reveals altered TGF-β sig-\nnalling and ECM remodelling in UF endometrium –\nThe impact offibroids on endometrial function has been reviewed by Ikhena\nand Bulun75.E l e v a t e dT G F -β3 secretion from fibroids is implicated in dis-\nrupting wound healing and coagulation pathways, potentially contributing\nto HMB\n76. To explore the molecular and cellular differences between UF and\nhealthy endometrium, we applied single-cell RNAseq (sc-RNAseq) on\nsecretory-phase endometrial samples from UF patients with HMB (n =4 ) ,\nintegrating them with healthy secretory-phase endometrium\n77 (n = 5).\nFollowing batch correction, quality control and cell annotation (Supple-\nmentary Figs. 8, 9 and Fig. 4b), we identi fied 4 major cell types, further\nsubdivided into 10 cell clusters (Fig.4a, b), including lymphatic endothelial\ncells, macrophage, and dendritic cells, with notable differences of cell\ncomposition between normal and UF tissues (Supplementary Fig. 10).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 7\n\nTo investigate cellular communication networks, we performed\nCellChat52 for ligand-receptor interaction analysis. We observed strikingly\nincreased cross-talks between UF endometrial cell clusters, compared to\nhealthy controls (Fig. 4c). Among the enriched signalling pathways ( p-\nvalue < 0.05), signalling by TGF-β superfamily, was markedly upregulated\n(Fig. 4d), with higher expression of TGF-β-associated receptors, including\nTGFBR1, TGFBR2, BMPR1A, BMPR1B, BMPR2 and ACVR1 in UF\nendometrium (average log\n2FC > 1.3, padj <0.05; Fig. 4e). Given the\nestablished elevation of TGF- β in fibroid78,79, these results suggest that\nfibroid-derived TGF-β ligands may contribute to aberrant signalling in\nsurrounding uterine tissues, potentially exacerbating HMB and ECM\nremodelling.\nNegative associated with HMB\nPositive associated with HMB\nNPTN\nPRELP\nS100A6\nSPEG\nPRUNE2\nVTN\nACTN1\nEMILIN1\nMXRA7\nANXA3\nHSPG2\nCD59\nNCAM1\nNEXN\nAOC3\nPDLIM5\nAHNAK\nCLU\nANG\nFERMT2\nCSDC2\nCAVIN1\nEPS8\nAGR2\nPAPSS1\nTMOD2\nMSRB3\nBNC2\nFILIP1\nMCAM\nDCN\nSFRP4\nMYH9\nCAPS\nRBPMS2\nDES\nPTGIS\nPGM5\nLPP\nSOD3\nMYH14\nMATN2\nNFIX\nPARVA\nVCAN\nSTOM\nLMOD1\nCOL6A2\nPODN\nDTNA\nEZR\nMYL9\nEHD2\nHSPB6\nANXA6\nJCAD\nRBP7\nTLN1\nTIMP3\nCAVIN2\nSNCG\nAKAP12\nPPP1R12C\nCNN1\nDMD\nAHNAK2\nFLNA\nHMGB3\nITGA1\nSYNPO\nJPH2\nTNS1\nGAS1\nHMGA1\nSVIL\nMMRN2\nDNAJB5\nSLC9A3R1\nCCDC9B\nASRGL1\nHSPB8\nADGRE5\nCSPG4\nCDH13\nMAP1B\nMFGE8\nABI3BP\nIGFBP5\nSORD\nCA2\nTGFB1I1\nFSCN1\nMYH11\nECM1\nPLIN4\nAngiogenesisAngiogenesisECM\norganisation\nECM\norganisation\nWound healingWound healing\nMAT2A\nPOLR2M\nAASDHPPT\nFLOT2\nSYF2\nMRPL1\nPPIL3\nTPT1\nRBM4\nYLPM1\nTF MTIF3POLDIP3\nLMAN2\nDHX38\nPPIG\nNFKB1\nTRA2A\nRNA metabolic\nprocess\nRNA/ mRNA\nsplicing\nNegative-associated with HMB/ MED12-mut fibroid)\nPositive-associated with HMB/ MED12-mut fibroid)\nTop Enriched Pathway\nFactor 7\n01234\n−log(p−value)\n−log(p−value)\npeptidyl−proline modification (GO:0018208)\nprotein peptidyl−prolyl isomerization (GO:0000413)\nmRNA splicing, via spliceosome (GO:0000398)\nRNA splicing, via transesterification reactions with\nbulged adenosine as nucleophile (GO:0000377)\nmRNA processing (GO:0006397)\nTop Enriched Pathway\ndermatan sulfate biosynthetic process (GO:0030208)\nnegative regulation of cell population proliferation\n(GO:0008285)\nregulation of muscle contraction (GO:0006937)\nsupramolecular fiber organization (GO:0097435)\nglycosaminoglycan catabolic process (GO:0006027)\nhomotypic cell−cell adhesion (GO:0034109)\nplatelet aggregation (GO:0070527)\nplasma membrane organization (GO:0007009)\nnegative regulation of cellular process (GO:0048523)\nmuscle contraction (GO:0006936)\n0246\nFactor 7\nFactor 2\nFactor 1\nRNAseq\nProteomics\n0\n5\n10\n15\n20\nVariance (%)\n0\n1\n2\n3\n-log10\npadj\nBatch\nFibroid  presence (UF/ non-UF)\nFibroid_MED12 UF\nFibroid_AHR rs2066853\nFibroid_FH rs6673988\nHMB_status\nHormone_past\nHormone_current\nFibroid_COL4A6 rs6622312\nFactor 1actomyosin structure organization (GO:0031032)\nplatelet degranulation (GO:0002576)\nplasma membrane repair (GO:0001778)\nchondroitin sulfate catabolic process (GO:0030207)\nregulated exocytosis (GO:0045055)\nHMB (Yes/No)\nFibroid (UF/non-UF)\nFibroid (UF/non-UF)\nHormone past\n(Yes/No/Unknown)\nHormone current\n(Yes/No/Unknown)\nHMB (Yes/No)\n−1\n0\n1\n2Factor value\nMED12\n(wt/mut/no fibroid)\nCOL4A6_rs6622312\n(wt/mut/no fibroid)\nAHR_rs2066853\n(wt/mut/no fibroid)\nFactor Factorr 7Factor 1 Factor 7Factor\n2\n−1\n0\n1\n2\nFactor 1\nHMB\nHMB No\nHMB\nHMB No\nFactor 7\n−0.2\n−0.1\n0.0\n0.1\n0.2\nFactor 1\nYes unknownNo\n−1\n0\n1\n2\n Past\nHormone treatment\nFactor 7\n−0.3\n−0.2\n−0.1\n0.0\n0.1\n0.2\nNo Fibroid\nRef wt\nvariant\nCOL4A6_rs6622312\nFibroid\nFactor 2\n−0.5\n0.0\n0.5\n1.0\nFibroid_MED12UF\nmut\nWT\nNo fibroid\nFibroid presence\nFactor 7\n−0.2\n−0.1\n0.0\n0.1\n0.2\nwt Fibroid\nMED12- UF\nFibroid\n−0.3\nNo Fibroid\nmut Fibroid\na b\nc\nd\ne\nf\ng\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 8\n\nApart from TGF- β receptors,T G F B 2expression was also notably\nelevated in UF endometrium (average log2FC > 3.0, padj <0.05), compared\nto healthy controls. As shown in Fig.4f, TGF-β2-mediated signalling via the\ndimer of the type I TGF-β receptor (TGFBR1) and ACVR1 revealed that in\nhealthy endometrium, signalling was primarily restricted to stromal clusters\n(p-value < 0.05), whereas in UF endometrium, it was widespread across\nmultiple cell types, indicating differences in TGF-β signalling. In addition,\nabnormal signalling pathways, including effectors such as collagen, laminin\nand fibronectin (FN1) were observed in UF endometrium and myome-\ntrium, compared to normal tissues\n77,80 (Supplementary Figs. 11, 12). This\nalso underscores alterations in ECM composition and basement membrane\narchitecture, which potentially compromise tissue homoeostasis and may\ncontribute to UF-associated pathophysiology.\nTGF-β signalling in THESC cells induces alternative splicing\nOur findings from bulk short-read RNAseq experiments suggested that\nTGF-β signalling induces alternative splicing changes in uterine tissues. To\ninvestigate the hypothesis that transcript isoform shifts are triggered by\nTGF-β in the endometrium, we treated the hTERT-immortalized human\nendometrial stromal cell line (THESC) with TGF-β during in vitro decid-\nualization and monitored transcript-level changes using Nanopore long-\nread RNA sequencing (Fig.5). This approach enabled precise determination\nof transcript isoforms. Consisten t with a short-read (Illumina) THESC\ndataset, differentially expressed genes (padj < 0.05, absolute log\n2FC ≥ 1.5)\nwere enriched in pathways related to cell cycle regulation and chromosome\nsegregation (Supplementary Fig. 13). The pro-fibrotic effects of TGF-β are\nmediated through both SMAD-dependent and non-canonical MEK/ERK\nsignalling pathways\n81,82. Prior studies have shown that blocking MEK/ERK\ncan attenuatefibroid cell proliferation and ECM production, suggesting that\nERK activation is required for certain TGF- β-mediated effects in fibroid\npathology83,84. Given that aberrant ECM accumulation and dysregulated\nangiogenesis are key contributors tofibroid-associated HMB, we applied a\nMAPK/ERK kinase (MEK) inhibitor (MEKi)59 to determine whether TGF-\nβ-mediated signalling relevant to these processes was dependent on MEK/\nERK activation.\nTo identify and quantify transcripts isoforms in the long-read dataset,\nwe employed Talon54 for transcript annotation and quantification, followed\nby Swan 55 for differential isoform expre ssion analysis. Notably, TGF- β\ntreatment during decidualization led to DTU events (p < 0.05) compared to\nDMSO, particularly in genes involved in mRNA processing and splicing,\nsuch as the hnRNP family\n85–87 (HNRNPA1, HNRNPA2B1, HNRNPC,\nHNRNPK, HNRNPR, HNRNPU), RNA-binding proteins (RBM4, RBM39),\nVEGFA-VEGFR2 signalling pathways , and hereditary leiomyomatosis\n(Supplementary Fig. 14a). Similar pathways were enriched when comparing\nco-treatment with TGF-β and MEKi to TGF- β treatment alone during\ndecidualization, indicating that TGF-β-driven transcriptome reprogram-\nming is largely achieved through RNA metabolic process and mRNA\nsplicing (Supplementary Fig. 14b).\nA ss h o w ni nF i g .5,T G F -β altered transcript isoform ratios in multiple\ngenes. For instance,HNRNPA2B1exhibited a shift from 100%A2B1-202to\na5 0 : 5 0r a t i oo fA2B1-202 and A2B1-206 upon TGF-β treatment (Fig. 5a,\nmiddle panel). Given that A2B1-206 is an intron-retained, non-protein-\ncoding transcript, this shift sugge sts potential downregulation of\nHNRNPA2B1. Similarly, we detected 10 HNRNPC transcript isoforms\n(Fig. 5a, bottom panel), including HNRNPC-206 (ENST00000553444), a\nnon-protein coding variant, while other isoforms encode structurally dis-\ntinct proteins, suggesting functional changes due to transcript switching.\nThesefindings indicate that the functions of hnRNP family are regulated via\nalternative transcript usage, subsequently further influencing mRNA spli-\ncing and processing.\nIn addition to splicing-related genes, alternative splicing in ECM-\nassociated genes was observed (Fig. 5b). Fibronectin-1 (FN1), a key ECM\nglycoprotein, mediates cell adhesion,integrin signalling, and growth factor\nbinding (including TGF-β interactions)\n88–90. With distinct domain com-\npositions, FN1 isoforms display different ligand-binding affinity, dimer-\nization, solubility, and fibrillogenesis88,89. Among the FN1 transcripts\nidentified (Fig.5b, upper panel), three are protein-coding.FN1-208encodes\na 73 kDa N-terminal protein, FN1-213 encodes a 121 kDa central/C-\nterminal protein, and FN1-207 encodes a 239 kDa full-length isoform\nlacking EDA and EDB regions. These isoforms may exert differential effects\nin ECM organisation. Additionally, periostin (POSTN), a secreted ECM\nglycoprotein involved in fibrosis and tumour progression\n91, exhibited\nalternative splicing between exon 17 and exon 21 (Fig. 5b, bottom panel),\nconsistent with its differential expression in normal and diseased tissues91.\nTo examine potential effects on HMB by blocking TGF-β or MAPK\npathways, we further tested MEK and ACVR1 (TGF-β receptor) inhibition\nin an in vivo mouse menstruation model 56. This system mimics primate\nmenstrual cycles, where progesterone withdrawal induces menstrual-like\nbleeding in ovariectomised, hormone-primed mice. MEK or ACVR1\ninhibition significantly reduced uterine bleeding (Fig. 5c), supporting the\nhypothesis that UF-associated growth factors affect endometrium phy-\nsiology that potentially contributing to HMB.\nOverall, our findings in both decidualized THESC cell line and\nendometrium from UF patients reveal that TGF- β signalling alters tran-\nscript usage in genes involved in mRNA splicing and ECM organisation.\nThese alternative splicing events may underlie key pathological changes in\nUF, contributing to endometrial dysfunction and heavy menstrual bleeding.\nDiscussion\nThe molecular mechanism linking UF s to HMB remains poorly under-\nstood, limiting targeted treatment o ptions while current treatments pri-\nmarily aim on reducing menstrual blood loss. UF growth is a female sex-\nsteroid hormone-dependent process, accordingly therapeutic interventions\nfor HMB have often focused on steroid hormones, oestrogen and proges-\nterone, including selective progesterone receptor modulators (SPRM) such\nas Ulipristal acetate (UPA) and gonadotropin-releasing hormone (GnRH)\nFig. 2 | Integrated analysis of endometrium using multi-omics, including tran-\nscriptomics, proteomics, and targeted genomic sequencing. a Left: The relative\ncontribution of the transcriptomic and proteomic datasets to MOFA-inferred fac-\ntors, expressed as the percentage of explained variance, with intensity represented in\nblue. Right: Correlation of factor variance with clinical and genetic parameters,\nquantified by -log\n10 (adjusted p-value) and visualised in red. Parameters include\nexperimental batches (n = 3), fibroid presence (UF vs non-UF), hormone treatment\n(past or current), heavy menstrual bleeding (HMB) status, and fibroid-associated\nmutations: canonical MED12 UF mutations, COL4A6 rs6622312, AHR rs2066853\nand FH rs6673988. b Scatter plots illustrating the differentiation of samples based on\nkey clinical and genetic parameters, including HMB (Yes, n = 10; No, n = 21),\nhormone past (prior hormone treatment: Yes, n = 11; No, n = 16; Unknown, n = 4),\nhormone current (treatment at time of surgery: Yes, n = 8; No, n = 20; Unknown,\nn = 3), fibroid (UF, n = 23; non-UF, n = 8), MED12 UF mutations (wt, n = 14; mut, n\n= 9; non-fibroid, n = 8), COL4A6 rs6622312 (wt, n = 12; mut, n = 11; non-fibroid,\nn = 8), and AHR rs2066853 (wt, n = 16; mut, n = 7; non- fibroid, n = 8). MOFA factor\nvalues represent the relative positioning of samples, with larger absolute values\nindicating stronger associations. c Boxplots showing the distribution of sample\ngroups across MOFA factors 1, 2, and 7, revealing variance within these factors. The\ncentre line represents the median; boxes represent the interquartile range (IQR), and\nwhiskers extend to 1.5 times of IQR. d Gene ontology (GO) enrichment analysis\nhighlighting pathways of features contributing to Factor 1 in both omics (FDR <\n0.1). e STRING network diagrams elucidating the interactions among features\nassociated with Factor 1 in both modalities (absolute loading weight higher than 0.3).\nThe loading weight of each feature was identi fied by MOFA using Bayesian fra-\nmework and sparsity-induced priors, different from classical regression using\np-values for significance. Only relevant features have non-zero loading weight. f GO\nenrichment analysis of features contributing to Factor 7 in both omics (FDR < 0.1).\ng STRING network diagrams of features associated with Factor 7 in both modalities\n(absolute loading weight higher than 0.3).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 9\n\nagonist therapy3–5. Due to their side effect profiles, non-hormonal therapies\nthat efficiently and safely target HMB are highly desirable. Our study pro-\nvides insights into the molecular mechanism underlying uterine fibroid\n(UF), particularly in relation to heavy menstrual bleeding (HMB). By\napplying the multi-omics analysis of transcriptomics, proteomics, and\ngenomics, in addition to single cell RNAseq (sc-RNAseq) analysis and\ndifferential transcript usage (DTU) analysis, we identified alternative tran-\nscript usage, TGF-β signalling and ECM dysregulation as key molecular\nalterations that contribute to fibroid pathogenesis and endometrial\ndysfunction.\nOur targeted sequencing approach reveals, in contrast to prior reports\nsuggesting thatMED12and HMGA2mutations are present in ~90% of UFs,\na lower frequency of these mutations (<50% of cases) in our cohort. The\nreason for the discrepancy is unknownbut the data may point to ethnic and\nregional differences in genomic aberrations found in UFs\n92,93. Instead, we\nobserved a higher prevalence of AHR missense mutations, and COL4A6\nENST00000295754 (encoded protein 567 aa)\nENST00000552516 (encoded protein 507 aa)\nENST00000374994 (encoded protein 503 aa)\nENST00000374990 (encoded protein 426 aa)\nTGFBR2 TGFBR3TGFBR1\nEstimated proportions\nTGFBR2\nTGFBR1\nENST00000533089 (nonsense mediated decay)\nENST00000212355 (encoded protein 851 aa)\nTGFBR3\n0.0\n0.4\n0.8\nENST00000359013 (encoded protein 592 aa)\nWT\nmut\nmut\nWT\nmut\nWT\nTGFBR2\nENST00000359013(encoded protein 592aa)ENST00000295754(encoded protein 567aa)\n0.00\n0.25\n0.50\n0.75\n1.00\nProportions\n0.00\n0.25\n0.50\n0.75\n1.00 TBXA2R\nENST00000589966(encoded protein 259aa)ENST00000375190(encoded protein 343aa)\n0.25\n0.50\n0.75\nENG\nENST00000480266(encoded protein 476aa)ENST00000373203(encoded protein 658aa)\nPDE1A\n0.0\n0.2\n0.4\n0.6\n0.8\nNRP1\n0.00\n0.25\n0.50\n0.75\n1.00\nENST00000351439(encoded protein 519aa)ENST00000410103(encoded protein 535aa)ENST00000435564(encoded protein 545aa)\n0.2\n0\n0.4\n0.6\nHNRNPR\nENST00000476660(CDS not defined)ENST00000302271(encoded protein 633aa)ENST00000374612(encoded protein 633aa)ENST00000374616(encoded protein 636aa)\nProportions 0.2\n0\n0.4\n0.6\n0.8\nTGFBI\nENST00000508076(encoded protein 65aa)ENST00000442011(encoded protein 483aa)ENST00000514554(encoded protein 366aa)\nMED12 WT\nMED12 mut\nMED12_UF Patient\nNo\nHMB\nHMB symptom\nHNRNPL\nTranscripts\n0\n0.4\n0.3\n0.2\n0.1\n0.5\nENST00000600873(encoded protein 456aa)\nENST00000595804(retained intron)\nENST00000647557(encoded protein 626aa)ENST00000601449(encoded protein 530aa)\nENST00000597731(retained intron)\nProportions\nANGPT2ANGPT1\n0.0\n0.4\n0.8Estimated proportions\nENST00000297450 (encoded protein 497 aa)\nENST00000523120 (encoded protein 459 aa)\nENST00000629816 (encoded protein 495 aa)\nENST00000325203 (encoded protein 496 aa)\nENST00000517746 (encoded protein 498 aa)\nENST00000520052 (encoded protein 297 aa)\nANGPT1\nANGPT2\nWT\nmut\nmut\nWT\nMED12_UF Patient\nTop Enriched Pathways\nTop Enriched Pathways\nGO: Biological Process\n02468\n−log(p−value)\nstress−activated MAPK cascade (GO:0051403)\nmRNA−containing ribonucleoprotein complex\nexport from nucleus (GO:0071427)\nregulation of spindle organization\n(GO:0090224)\nregulation of RNA splicing (GO:0043484)\nmRNA transport (GO:0051028)\ncellular response to transforming growth\nfactor beta stimulus (GO:0071560)\npost−translational protein\nmodification (GO:0043687)\nreceptor−mediated endocytosis\n(GO:0006898)\ntransforming growth factor beta receptor\nsignaling pathway (GO:0007179)\nextracellular matrix organization (GO:0030198)\ntranscription initiation from RNA polymerase\nIII promoter (GO:0006384)\ncellular response to DNA damage\nstimulus(GO:0006974)\nregulation of translation (GO:0006417)\nregulation of apoptotic process (GO:0042981)\ncellular protein modification process\n(GO:0006464)\nGO: Biological Process\n0.0 0.2 0.4 0.6\n−log(p−value)\nrRNA processing (GO:0006364)\nregulation of phosphorylation (GO:0042325)\nprotein localization to cell−cell junction\n(GO:0150105)\nribosome biogenesis (GO:0042254)\npositive regulation of cellular protein metabolic\nprocess (GO:0032270)\nnegative regulation of vascular associated smooth\nmuscle cell differentiation (GO:1905064)\nbranching morphogenesis of an epithelial tube\n(GO:0048754)\nregulation of vascular associated smooth muscle\ncell differentiation (GO:1905063)\npeptide antigen assembly with MHC protein complex\n(GO:0002501)\na\nb c\nd\ne\nENST00000374875(encoded protein 735aa)ENST00000374867(encoded protein 923aa)\nFig. 3 | Comparative analysis of transcript usage in active endometrium from\npatients with heavy menstrual bleeding or MED12-mutated fibroids. a Boxplots\ndisplaying the expression of differentially used transcript variants in the active\nendometrium of UF patients with heavy menstrual bleeding (HMB, n = 8; coloured\nin pink-orange) compared to non-HMB patients ( n = 7; grey). The centre line\nrepresents the median, while the lower and upper hinges correspond to the 25\nth and\nthe 75th percentiles. b, c Bar plots of enriched pathways associated with genes\nexhibiting differential transcript usage, identi fied using DRIMSeq ( p-value < 0.05)\nand stageR (OFDR < 0.05). b Pathways enriched in HMB versus non-HMB endo-\nmetrium. c Pathways enriched in endometrium from MED12-mutant versus\nMED12 wild-type (WT) fibroid patients. d Boxplots showing the expression of\ndifferentially used transcript variants in active endometrium from MED12-mutant\nUF patients ( n = 10; pink-orange) versus MED12 WT (n = 5; grey). e Ribbon plots\nillustrating transcript usage shifts between MED12 WT and MED12-mutant con-\nditions, highlighting dynamic usage patterns across transcript variants of individual\ngenes. Each transcript per gene is represented by a distinct colour.\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 10\n\n−10\n−5\n0\n5\n10\n15\n−10 −5 0 5 10\nUMAP_1\na b\nc d\ne f\nUMAP_2\nEndothelial\nLymphatic_EC - 9 , 21\nA r t e r y _ E C-2 0\nImmune:\nMacrophage - 23\nNK/T cells - 16\nDendritic cells - 22\nStromal\nDES+ACTA2+FAP+\nRGS5+CSPG4+ - 7, 14, 18\nACTA2+FAP+ - 0, 1, 10\nFAP+ - 11, 19\nEpithelial\nEpi:unciliated - 2, 3, 4, 5,\n6, 8, 12, 13,1 5, 17\nEpi:ciliated - 6, 24\nMacrophage\nNK/T cells\nDendritic cells\nCiliated Epithelial\nUnciliated Epithelial\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nLymphatic_EC\nArtery_EC\nMacrophage\nNK/T cells\nDendritic cells\nCiliated Epithelial\nUnciliated Epithelial\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nLymphatic_EC\nArtery_EC\n0\n10\n20\n0 15\ndifferential interactions in\nUF Endometrium\nRelative values−1\n0\n1\n2\n3\n4\nSources (Sender)\n2\n6\n4\n5\n16\n0\n9\n19\n1\n3\n22\n17\n8\n14\n7\n18\n10\n12\n15\n20\n2124 25\n13\n11\n23\n−1.0\n−0.5\n0.0\n0.5\n1.0\nExpressionMacrophage\nImmune cluster\nNK/T cells\nDendritic cells\n−1.0\n−0.5\n0.0\n0.5\n1.0\nExpression\nLUM\nCOL6A3\nDCN\nDES\nCNN1\nACTA2\nBGN\nMCAM\nPDGFRB\nCSPG4\nSUSD2\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nStromal cluster\nFCER1A\nHLA-DQB1\nGNLY\nNKG7\nITK\nCD2\nMSR1\nMRC1\nKIT\nTPSAB1\nHLA-A\nCD38\nCD74\nExpression\nEndothelial cluster\nLymphatic_EC\nArtery_EC\nMMRN1\nPROX1\nPKHD1L1\nSEMA3D\nRELN\nKLHL4\nDKK2\nIGFBP3\nFBLN5\nSERPINE2\nGJA5\nCXCL12\nBTNL9\nRGCC\nADGRF5\nKIAA1217\nSELP\nCOL15A1\nZNF385D\nEBF1\nTSHZ2\nCPXM2\nTPD52L1\nPDE7B\nACKR1\nITM2A\nCCL14\nCLU\nHLA−DRB1\nCD74\nRAMP3\nMALAT1\nNEAT1\nXIST\nMACF1−0.4\n0.0\n0.4\nMacrophage\nNK/T cells\nDendritic cells\nCiliated Epithelial\nUnciliated Epithelial\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nLymphatic_EC\nArtery_EC\nMacrophage\nNK/T cells\nDendritic cells\nCiliated Epithelial\nUnciliated Epithelial\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nLymphatic_EC\nArtery_EC\n0 0.04\nNormal Endometrium\nSources (Sender)\n0\n0.08\nCommunication Prob.\n0\n0.005\n0.01\n0.015\n0.02\nMacrophage\nNK/T cells\nDendritic cells\nCiliated Epithelial\nUnciliated Epithelial\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nACTA2+FAP+\nFAP+\nLymphatic_EC\nArtery_EC\n0 0.3\nUF Endometrium\n0\n0.2\nTGF-beta signaling\nEC:Artery\nEC:Lymphatic\nEpi:ciliated\nEpi:\nunciliated\nImmune:\nDC\nImmune:\nMacrophage\nImmune:\nNK/T cells\nStromal:\nACTA2+FAP+\nStromal:\nDES+ACTA2+FAP+\nRGS5+CSPG4+\nStromal: FAP+\nUF\nBMPR1B\nACVR1\nBMPR2\n4\n3\n3\nBMPR1A\nACVR2A\nACVR1B\n4\n4\n5\n3\nTGFBR2\nTGFB1\n4\n6\nTGFB2\nTGFBR1\n5\nArtery endothelial\nLymphatic endothelial\nCiliated epithelial\nUnciiliated epithelial\nDendritic cellsMacrophageNK/\n T cells\nACTA2\n+FAP\n+\nFAP\n+\nDES\n+ACT\nA2\n+FAP\n+\nRGS5\n+CSPG4\n+\nNormal UF\nEC:Artery\nEC:Lymphatic\nEpi:\nciliated\nEpi:\nunciliated\nImmune:\nDC\nImmune:\nMacrophage\nImmune:\nNK/T cells\nStromal:\nACTA2+FAP+\nStromal:\nDES\n+ACTA2+FAP+\nRGS5+CSPG4+\nStromal:FAP+\nTGFB2 − (ACVR1+TGFBR1)\nNormal\nFig. 4 | Single-cell analysis of endometrium from UF patients with heavy men-\nstrual bleeding compared to healthy controls. a UMAP of the integrative single-\ncell dataset of UF ( n = 4) and healthy endometrium ( n = 5). Colours represent dis-\ntinct cell subclusters within major cell types. b Heatmaps exhibiting average\nexpression of canonical marker genes used for cell type annotation in stromal (upper\nleft), immune (upper right) and endothelial (bottom) clusters. c Heatmap of dif-\nferentially enriched cell-cell interactions in UF endometrium compared to healthy\ncontrols. Relative values of interaction strength is indicated by a gradient from blue\n(low) to red (high). d Heatmap displaying TGF- β signalling across cell clusters in\nnormal (left) and UF (right) endometrium. e Violin plots illustrating the expression\nof ligands and receptors involved in TGF-β signalling in normal (blue) and UF (red)\nendometrium.f Circle plots showing inferred TGFB2-(ACVR1 + TGFBR1) sig-\nnalling among different cell types in normal (top) and UF (bottom) endometrium.\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 11\n\ninsertion-deletion and frameshift variants. Given that ECM dysregulation is\na hallmark of UF, the identification ofCOL4A6variants further underscores\nthe functional impacts on ECM remodelling. In addition to hormone reg-\nulation, key mechanisms that cont ribute to ECM remodelling in UFs\ninclude Rho and ERK/p38 related mechanotransduction, nuclear location of\nYAP/TAZ, and growth factors such as TGF-β,E G F ,a n dI G F - 1\n94–96.\nSupporting this, our multi-omics analysis confirmed upregulation of\nkey ECM components, includingCOL1A1, COL3A1and VCAN,c o n s i s t e n t\nwith previous studies demonstrating excessive collagen synthesis in\nUFs25,94,97–100.C o l l a g e nfibrils for example, were found shorter and more\ndisordered in UFs, in addition to the altered ratio of collagen type I/III101.\nMoreover, sc-RNAseq revealed elevat ed receptor-ligand interactions in\ncollagen, laminin, and fibronectin-1 signalling in UF endometrium and\nmyometrium, suggesting that ECM remodelling extends beyond fibroid\nitself to the surrounding uterine tissues 31,102.T h e s efindings reinforce the\nhypothesis that targeting ECM-rel ated pathways may offer therapeutic\npotential in treating UF and its associated symptoms94,95,103–107.\nOur study also highlights RNA processing and alternative splicing as\ncritical contributors to endometrial dysfunction in UF patients. Alternative\nsplicing plays a crucial role in protein diversity and has been linked to\nvarious diseases, including cancer\n108–110. Our multi-omic analysis identified\nlatent factors that correlates with HMB, hormone treatment, and fibroid\npresence with certain driver mutations, emphasizing the broad impact of UF\non endometrial physiology. We found that RNA metabolic processes and\nsplicing-related genes were noticeably dysregulated, implicating that aber-\nrant transcript usage may contribute to UF-associated HMB.\nFurther DTU analysis revealed alternative splicing in genes involved in\nblood vessel morphogenesis ( TGFBR2, ENG,a n d NRP1), prostaglandin\nsynthesis (TBXA2R, PTGES), and hormone signalling ( PGR, FGF7 and\nFGFR2). DTU in splicing-related genes ( HNRNPR, HNRNPL)f u r t h e r\nunderscores the potential disruption of splicing regulation in UF-associated\nendometrial pathology.\nNotably, the TGF-β type II receptor emerged as a key regulator, with an\naltered balance between its two isoforms, TβR-II and TβRII-B, which binds\nTGF-β I/III or TGF-β II, respectively\n72,73.O u rfindings suggest a shift toward\nthe dominant expression of T βRII-B in UF endometrium, potentially\ninfluencing TGF-β ligand specificity and downstream signalling effects.\nThese findings suggest that alternative splicing in UF endometrium may\nalter TGF-β signalling dynamics, further compromising endometrial tissue\nhomoeostasis, ECM remodelling andfibrotic processes.\nTGF-β signalling111,112 is a known regulator of alternative splicing,\nacting through pathways such as SMAD and PI3K/Akt/SRPK1 113–118 to\ninfluence exon inclusion and exclusion. Our sc-RNAseq analysis revealed\nTGF-β signalling is strikingly upregulated in UF endometrium, with ele-\nvated expression of TGF-β receptors. Given the well-established elevation of\nTGF-β levels in UF tissues, ourfindings suggest thatfibroids may serve as a\nsource of TGF-β ligands, which in turn influence alternative splicing and\ntranscript expression profile in endometrium.\nTo validate the role of alternative splicing in endometrial physiology,\nwe examined transcript isoform changes in vitro using TGF- β treated\nTHESC cells during decidualization.Long-read sequencing analysis iden-\ntified DTU in genes regulating RNA splicing including hnRNP family,\nRBM4 and RBM39, ECM organisation like FN1, POSTN, and immune\nresponse like CD59119–124. We showed a shift in isoform ratios for HNRNP\ngenes, FN1,a n dPOSTN, suggesting that TGF-β signalling may affect ECM\nDecidualization\nDecidualization\nPercentage of\nHNRNPA2B1 isoform\nPercentage of\nHNRNPA1 isoform\nPercentage of\nHNRNPC isoform\nDecidualization\n0\n20\n40\n60\n80\n100\nMEKi\nHNRNPC-205 HNRNPC-211 HNRNPC-230\nHNRNPC-219 HNRNPC-222 HNRNPC-201\nHNRNPC-208 HNRNPC-214 HNRNPC-206\n0\n20\n40\n60\n80\n100\n0\n20\n40\n60\n80\n100\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nMEKi\nHNRNPA2B1-202 HNRNPA2B1-206\nHNRNPA2B1-201\nCtrl DMSO    TGF-b e t a  TGF-beta\n  +MEKi\nMEKi\nHNRNPA1-203 HNRNPA1-202\n5 kb\n5 kb\nTranscript Model of HNRNPA1Transcript Name\nHNRNPA1-203\nHNRNPA1-202\nTranscript Model of HNRNPA2B1\n50 kb\nTranscript Model of HNRNPC\nTranscript Name\nHNRNPA2B1-202\nHNRNPA2B1-206\nHNRNPA2B1-201\nTranscript Name\nHNRNPC-205\nHNRNPC-211\nHNRNPC-230\nHNRNPC-219\nHNRNPC-222\nHNRNPC-201\nHNRNPC-208\nHNRNPC-214\nHNRNPC-206\nPercentage of\nPOSTN isoform\nPercentage of\nCD59 isoform\nPercentage of\nFN1 isoform\nDecidualization\nDecidualization\n0\n20\n40\n60\n80\n100\n0\n20\n40\n60\n80\n100\nMEKi\nDecidualization\nMEKi\nCD59-203 CD59-211 CD59-201\nCD59-209 CD59-205 CD59-202\n0\n20\n40\n60\n80\n100\nMEKi\nFN1-213 FN1-207\nFN1-208 FN1-227 FN1-225\nPOSTN-209 ENCODE_hg_v29T000242753\nPOSTN-201 POSTN-202\nPOSTN-210 POSTN-204\nTranscript Model of FN1Transcript Name\nFN1-208\nFN1-227\nFN1-225\nFN1-213\nFN1-207\nTranscript Model of CD59Transcript Name\nCD59-203\nCD59-211\nCD59-201\nCD59-209\nCD59-205\nCD59-202\nTranscript Model of POSTNTranscript Name\nPOSTN-209\nENCODE_hg_\nv29T000242753\nPOSTN-201\nPOSTN-202\nPOSTN-210\nPOSTN-204\nBlood loss [µl]\nBlood loss [µl]\nVehicle\nVehicle BAY MEKi\n(0.5 mg/kg/d)\n-40%* -81%****\nTP-0184\n(15 mg/kg/d)\n0\n25\n50\n75\n100\n-20\n0\n20\n40\n60\n80\n100\na\nb\nc\nFig. 5 | The effect of TGF- β on endometrial homoeostasis in vitro and in vivo.\na, b Alternative transcript usage induced by TGF- β treatment in decidualized\nTHESC cell line. a Members of the heterogeneous nuclear ribonucleoprotein\n(hnRNP) family: HNRNPA1, HNRNPA2B1, and HNRNPC. b ECM-related genes\n(FN1, POSTN) and the immune-related gene CD59. Sample size per group is 3.\nc Effect of TP-0184 (ACVR inhibitor; left panel) and BAY MEKi (MEK inhibitor;\nright panel) on menstrual-like bleeding in a murine model. Two independent in vivo\nexperiments were conducted to investigate the effects of TP-0184 (an ACVR inhi-\nbitor) and BAY-533 (a MEK inhibitor); the sample size for each group is 10. Both\ntreatments showed a significant reduction in total uterine blood loss. Blood loss was\nquantified via alkaline elution of tampons and corrected for background levels. Data\nrepresents the mean with standard deviation from ten experiments per treatment\ngroup. Statistical signi ficance was assessed using Student ’s t-test (*p < 0.05;\n****p < 0.0001).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 12\n\nremodelling and UF progression via directly influencing alternative splicing\nfactors and subsequent splicing events.\nThe MEK/ERK (MAPK) pathway plays a critical role in uterinefibroid\npathophysiology, particularly in mediatingfibroid cell proliferation, extra-\ncellular matrix (ECM) deposition, and angiogenesis, all of which contribute\nto heavy menstrual bleeding (HMB). Several studies have demonstrated that\ngrowth factors highly expressed infibroids, such as TGF-β,I G F ,a n dP D G F ,\nactivate the MEK/ERK pathway, driving fibrotic and angiogenic changes\nthat disrupt endometrial homoeostasis\n83,84,95,125.O u rfindings that blocking\nthe TGF- β or MEK signalling cascade in murine menstruation model\nreduced blood loss indicates TGF- β-driven changes, particularly those\naffecting the ECM and vasculature, contribute to HMB through an ERK-\ndependent pathway.\nOur findings have noticeable implications for potential UF treatment\nstrategies. Given that ECM stiffness has been linked to alternative splicing\nthrough activation of Ser/Arg-rich spliceosome proteins\n126,t a r g e t i n gE C M\nremodelling and TGF-β-mediated splicing regulation may provide potential\ntherapeutic avenues. Current antifibrotic approaches, such as collagenase\ntreatment or inhibition of fibrotic gene expression, have been shown to\nreduce ECM density andfibroid cell proliferation94. Moreover, compounds\nsuch as epigallocatechin gallate (EGCG) from green tea, have been shown to\nreduce fibroid volume and improve HMB, potentially through targeting on\nfibrotic signalling pathways including TGF- β, β-catenin, JNK and AKT\npathways which are involved in fibrotic progression127. Further studies\nshould further explore the therapeutic potential of splicing modulators and\nantifibrotic agents in mitigating UF progression and associated symptoms.\nData availability\nAll raw and processed sequencing data associated with Figs. 1–5 and all\nSupplementary Figs. in this study are available in the NCBI ’sG e n e\nExpression Omnibus: bulk RNA Sequencing data (GSE199849) and single-\ncell RNA sequencing data (GSE220650) of patient samples applied to this\nstudy; Illumina short-read and ONT long-read RNA sequencing of in vitro\nTHESC decidualization (GSE261366). The mass spectrometry proteomics\ndata have been deposited to the Prot eomeXchange Consortium via the\nPRIDE\n128 partner repository with the dataset identi fier PXD051220. The\nsource data for the graphs in Figs.1–5 in the main manuscript can be found\nin the Supplementary Data 6.\nReceived: 27 May 2024; Accepted: 18 July 2025;\nReferences\n1. Baird, D. D., Dunson, D. B., Hill, M. C., Cousins, D. & Schectman, J.\nM. High cumulative incidence of uterine leiomyoma in black and\nwhite women: ultrasound evidence. Am. J. Obstet. Gynecol. 188,\n100–107 (2003).\n2. Gupta, S., Jose, J. & Manyonda, I. Clinical presentation of fibroids.\nBest. Pr. Res Clin. Obstet. Gynaecol. 22, 615–626 (2008).\n3. Moravek, M. B. & Bulun, S. E. Endocrinology of uterine fibroids:\nsteroid hormones, stem cells, and genetic contribution. Curr. Opin.\nObstet. Gynecol. 27, 276–283 (2015).\n4. Ali, M. et al. Progesterone signaling and uterine fibroid pathogenesis;\nmolecular mechanisms and potential therapeutics.Cells12, 1117 (2023).\n5. Ploumaki, I., Macri, V. I., Segars, J. H. & Islam, M. S. Progesterone\nsignaling in uterine fibroids: Molecular mechanisms and therapeutic\nopportunities. Life Sci. 362, 123345 (2025).\n6. Stewart, E. A. Clinical practice. Uterine fibroids. N. Engl. J. Med.372,\n1646–1655 (2015).\n7. Jacobson, G. F., Shaber, R. E., Armstrong, M. A. & Hung, Y. Y.\nHysterectomy rates for benign indications. Obstet. Gynecol. 107,\n1278–1283 (2006).\n8. Cardozo, E. R. et al. The estimated annual cost of uterine\nleiomyomata in the United States. Am. J. Obstet. Gynecol. 206, 211\ne211–e219 (2012).\n9. Wegienka, G. et al. Self-reported heavy bleeding associated with\nuterine leiomyomata. Obstet. Gynecol. 101, 431–437 (2003).\n10. Cooper, K. G., Jack, S. A., Parkin, D. E. & Grant, A. M. Five-year\nfollow up of women randomised to medical management or\ntranscervical resection of the endometrium for heavy menstrual loss:\nclinical and quality of life outcomes. BJOG 108, 1222–1228 (2001).\n11. Goodman, A. Abnormal genital tract bleeding. Clin. Cornerstone 3,\n25–35 (2000).\n12. Hapangama, D. K. & Bulmer, J. N. Pathophysiology of heavy\nmenstrual bleeding. Women’s. Health 12,3 –13 (2016).\n13. Makinen, N. et al. MED12, the mediator complex subunit 12 gene, is\nmutated at high frequency in uterine leiomyomas. Science\n334,\n252–255 (2011).\n14. Mehine, M. et al. Characterization of uterine leiomyomas by whole-\ngenome sequencing. N. Engl. J. Med. 369,4 3–53 (2013).\n15. Elmlund, H. et al. The cyclin-dependent kinase 8 module sterically\nblocks Mediator interactions with RNA polymerase II. Proc. Natl.\nAcad. Sci. USA 103, 15788–15793 (2006).\n16. Cleynen, I. et al. HMGA2 regulates transcription of the Imp2 gene via\nan intronic regulatory element in cooperation with nuclear factor-\nkappaB. Mol. Cancer Res. 5, 363–372 (2007).\n17. Ono, M. et al. Paracrine activation of WNT/beta-catenin pathway in\nuterine leiomyoma stem cells promotes tumor growth. Proc. Natl.\nAcad. Sci. USA 110, 17053–17058 (2013).\n18. Stewart, E. A. et al. Uterine fibroids. Nat. Rev. Dis. Prim. 2, 16043\n(2016).\n19. Yang, C. E. et al. Aryl hydrocarbon receptor: From pathogenesis to\ntherapeutic targets in aging-related tissue fibrosis. Ageing Res. Rev.\n79, 101662 (2022).\n20. Mehine, M. et al. Integrated data analysis reveals uterine leiomyoma\nsubtypes with distinct driver pathways and biomarkers. Proc. Natl.\nAcad. Sci. USA 113, 1315–1320 (2016).\n21. Arslan, A. A. et al. Gene expression studies provide clues to the\npathogenesis of uterine leiomyoma: new evidence and a systematic\nreview. Hum. Reprod. 20, 852–863 (2005).\n22. Vanharanta, S. et al. Distinct expression pro file in fumarate-hydratase-\ndeficient uterine fibroids. Hum. Mol. Genet. 15,9 7–103 (2006).\n23. Christacos, N. C., Quade, B. J., Dal Cin, P. & Morton, C. C. Uterine\nleiomyomata with deletions of Ip represent a distinct cytogenetic\nsubgroup associated with unusual histologic features. Genes\nChromosomes Cancer 45, 304–312 (2006).\n24. Vanharanta, S. et al. 7q deletion mapping and expression pro filing in\nuterine fibroids. Oncogene 24, 6545–6554 (2005).\n25. Leppert, P. C., Catherino, W. H. & Segars, J. H. A new hypothesis\nabout the origin of uterinefibroids based on gene expression profi\nling\nwith microarrays. Am. J. Obstet. Gynecol. 195, 415–420 (2006).\n26. Zavadil, J. et al. Pro filing and functional analyses of microRNAs and\ntheir target gene products in human uterine leiomyomas. PLoS One\n5, e12362 (2010).\n27. Hodge, J. C. et al. Expression pro filing of uterine leiomyomata\ncytogenetic subgroups reveals distinct signatures in matched\nmyometrium: transcriptional profilingof the t(12;14) and evidence in\nsupport of predisposing genetic heterogeneity.Hum. Mol. Genet21,\n2312–2329 (2012).\n28. Cirilo, P. D. An integrative genomic and transcriptomic analysis\nreveals potential targets associated with cell proliferation in uterine\nleiomyomas. PLoS One 8, e57901 (2013).\n29. Ko, Y. A. et al. Extracellular matrix (ECM) activates beta-catenin\nsignaling in uterine fibroids. Reproduction 155,6 1–71 (2018).\n30. Jamaluddin, M. F. B. et al. Proteomic Pro filing of Human Uterine\nFibroids Reveals Upregulation of the Extracellular Matrix Protein\nPeriostin. Endocrinology 159, 1106–1118 (2018).\n31. Jamaluddin, M. F. B., Nahar, P. & Tanwar, P. S. Proteomic\ncharacterization of the extracellular matrix of human uterinefibroids.\nEndocrinology 159, 2656–2669 (2018).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 13\n\n32. Jamaluddin, M. F. B., Nagendra, P. B., Nahar, P., Oldmeadow, C. &\nTanwar, P. S. Proteomic analysis identifies Tenascin-C expression is\nupregulated in uterine fibroids. Reprod. Sci. 26, 476–486 (2019).\n33. Danecek, P. et al. Twelve years of SAMtools and BCFtools.\nGigascience 10, giab008 (2021).\n34. Cingolani, P. et al. A program for annotating and predicting the\neffects of single nucleotide polymorphisms, SnpEff: SNPs in the\ngenome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly\n6,8 0–92 (2012).\n35. McLaren, W. et al. The Ensembl variant effect predictor. Genome\nBiol. 17, 122 (2016).\n36. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold\nchange and dispersion for RNA-seq data with DESeq2. Genome\nBiol. 15, 550 (2014).\n37. Yu, G., Wang, L. G., Han, Y. & He, Q. Y. clusterPro filer: an R package\nfor comparing biological themes among gene clusters. OMICS 16,\n284–287 (2012).\n38. Wu, T. et al. clusterPro filer 4.0: A universal enrichment tool for\ninterpreting omics data. Innovations 2, 100141 (2021).\n39. Soneson, C., Love, M. I. & Robinson, M. D. Differential analyses for\nRNA-seq: transcript-level estimates improve gene-level inferences.\nF1000Res 4, 1521 (2015).\n40. Nowicka, M. & Robinson, M. D. DRIMSeq: a Dirichlet-multinomial\nframework for multivariate count outcomes in genomics. F1000Res\n5, 1356 (2016).\n41. Van den Berge, K., Soneson, C., Robinson, M. D. & Clement, L.\nstageR: a general stage-wise method for controlling the gene-level\nfalse discovery rate in differential expression and differential\ntranscript usage. Genome Biol. 18, 151 (2017).\n42. Bodenhofer, U., Bonatesta, E., Horejs-Kainrath, C. & Hochreiter, S.\nmsa: an R package for multiple sequence alignment. Bioinformatics\n31, 3997–3999 (2015).\n43. Dragicevic, M. B., Paunovic, D. M., Bogdanovic, M. D., Todorovic, S.\nI. & Simonovic, A. D. ragp: Pipeline for mining of plant\nhydroxyproline-rich glycoproteins with implementation in R.\nGlycobiology https://doi.org/10.1093/glycob/cwz072 (2019).\n44. Abramson, J. et al. Accurate structure prediction of biomolecular\ninteractions with AlphaFold 3. Nature 630, 493–500 (2024).\n45. Argelaguet, R. et al. Multi-Omics Factor Analysis-a framework for\nunsupervised integration of multi-omics data sets. Mol. Syst. Biol.\n14, e8124 (2018).\n46. Argelaguet, R. et al. MOFA +: a statistical framework for comprehensive\nintegration of multi-modal single-cell data.Genome Biol.21, 111 (2020).\n47. Kuleshov, M. V. et al. Enrichr: a comprehensive gene set enrichment\nanalysis web server 2016 update. Nucleic Acids Res. 44, W90\n–W97\n(2016).\n48. Amezquita, R. A. et al. Orchestrating single-cell analysis with\nBioconductor. Nat. Methods 17, 137–145 (2020).\n49. Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell\n184, 3573–3587.e3529 (2021).\n50. Germain, P. L., Lun, A., Garcia Meixide, C., Macnair, W. & Robinson,\nM. D. Doublet identi fication in single-cell sequencing data using\nscDblFinder. F1000Res 10, 979 (2021).\n51. Korsunsky, I. et al. Fast, sensitive and accurate integration of single-\ncell data with Harmony. Nat. Methods 16, 1289–1296 (2019).\n52. Jin, S. et al. Inference and analysis of cell-cell communication using\nCellChat. Nat. Commun. 12, 1088 (2021).\n53. Li, H. Minimap2: pairwise alignment for nucleotide sequences.\nBioinformatics 34, 3094–3100 (2018).\n54. Wyman, D. et al. A technology-agnostic long-read analysis pipeline\nfor transcriptome discovery and quanti fication. bioRxiv. https://doi.\norg/10.1101/672931 (2020).\n55. Reese, F. & Mortazavi, A. Swan: a library for the analysis and\nvisualization of long-read transcriptomes. Bioinformatics 37,\n1322–1323 (2021).\n56. Menning, A. et al. Granulocytes and vascularization regulate uterine\nbleeding and tissue remodeling in a mouse menstruation model.\nPLoS One 7, e41800 (2012).\n57. Cohen, P. E. & Milligan, S. R. Silastic implants for delivery of\noestradiol to mice. J. Reprod. Fertil. 99, 219–223 (1993).\n58. Hallberg, L. & Nilsson, L. Determination of Menstrual Blood Loss.\nScand. J. Clin. Lab Invest. 16, 244–248 (1964).\n59. Hartung, I. V. et al. Modular Assembly of Allosteric MEK inhibitor\nstructural elements unravels potency and feedback-modulation\nhandles. ChemMedChem 10, 2004–2013 (2015).\n60. Williams, A. J., Powell, W. L., Collins, T. & Morton, C. C. HMGI(Y)\nexpression in human uterine leiomyomata. Involvement of another\nhigh-mobility group architectural factor in a benign neoplasm.Am. J.\nPathol. 150, 911–918 (1997).\n61. Yatsenko, S. A. et al. Highly heterogeneous genomic landscape of\nuterine leiomyomas by whole exome sequencing and genome-wide\narrays. Fertil. Steril. 107, 457–466.e459 (2017).\n62. Schoenmakers, E. F. et al. Identi fication of CUX1 as the recurrent\nchromosomal band 7q22 target gene in human uterine leiomyoma.\nGenes Chromosomes Cancer 52,1 1–23 (2013).\n63. Medikare, V., Kandukuri, L. R., Ananthapur, V., Deenadayal, M. &\nNallari, P. The genetic bases of uterine fibroids; a review. J. Reprod.\nInfertil. 12, 181–191 (2011).\n64. Maybin, J. A. & Critchley, H. O. Menstrual physiology: implications\nfor endometrial pathology and beyond. Hum. Reprod. Update 21,\n748–761 (2015).\n65. Ragni, M. V. et al. Phase II prospective open-label trial of\nrecombinant interleukin-11 in women with mild von Willebrand\ndisease and refractory menorrhagia. Thromb. Haemost. 106,\n641–645 (2011).\n66. Magnay, J. L., O ’Brien, S., Gerlinger, C. & Seitz, C. Pictorial methods\nto assess heavy menstrual bleeding in research and clinical\npractice: a systematic literature review. BMC Women’s. Health 20,\n24 (2020).\n67. Brodsky, R. A. Paroxysmal nocturnal hemoglobinuria. Blood 124,\n2804–2811 (2014).\n68. Dickson, K. A. et al. Ribonuclease inhibitor regulates\nneovascularization by human angiogenin. Biochemistry 48,\n3804–3806 (2009).\n69. Mondal, M., Conole, D., Nautiyal, J. & Tate, E. W. UCHL1 as a novel\ntarget in breast cancer: emerging insights from cell and chemical\nbiology. Br. J. Cancer 126,2 4–33 (2022).\n70. Suzuki, T. et al. Pivotal Role of Ubiquitin Carboxyl-Terminal\nHydrolase L1 (UCHL1) in Uterine Leiomyoma. Biomolecules 13,\nhttps://doi.org/10.3390/biom13020193 (2023).\n71. Cox, J. & Mann, M. 1D and 2D annotation enrichment: a statistical\nmethod integrating quantitative proteomics with complementary\nhigh-throughput data. BMC Bioinforma. 13, S12 (2012).\n72. Rotzer, D. et al. Type III TGF-beta receptor-independent signalling of\nTGF-beta2 via TbetaRII-B, an alternatively spliced TGF-beta type II\nreceptor. EMBO J. 20, 480–490 (2001).\n73. del Re, E., Babitt, J. L., Pirani, A., Schneyer, A. L. & Lin, H. Y. In the\nabsence of type III receptor, the transforming growth factor (TGF)-\nbeta type II-B receptor requires the type I receptor to bind TGF-\nbeta2. J. Biol. Chem. 279, 22765–22772 (2004).\n74. Hirai, R. & Fijita, T. A human transforming growth factor-beta type II\nreceptor that contains an insertion in the extracellular domain. Exp.\nCell Res. 223, 135–141 (1996).\n75. Ikhena, D. E. & Bulun, S. E. Literature Review on the Role of Uterine\nFibroids in Endometrial Function. Reprod. Sci. 25\n,6 3 5–643\n(2018).\n76. Sinclair, D. C., Mastroyannis, A. & Taylor, H. S. Leiomyoma\nsimultaneously impair endometrial BMP-2-mediated\ndecidualization and anticoagulant expression through secretion of\nTGF-beta3. J. Clin. Endocrinol. Metab. 96, 412–421 (2011).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 14\n\n77. Wang, W. et al. Single-cell transcriptomic atlas of the human\nendometrium during the menstrual cycle. Nat. Med. 26, 1644–1653\n(2020).\n78. Arici, A. & Sozen, I. Transforming growth factor-beta3 is expressed\nat high levels in leiomyoma where it stimulates fibronectin\nexpression and cell proliferation. Fertil. Steril. 73, 1006–1011 (2000).\n79. Lee, B. S. & Nowak, R. A. Human leiomyoma smooth muscle cells\nshow increased expression of transforming growth factor-beta 3\n(TGF beta 3) and altered responses to the antiproliferative effects of\nTGF beta. J. Clin. Endocrinol. Metab. 86, 913–920 (2001).\n80. Paul, E. N. et al. Cysteine-rich intestinal protein 1 is a novel surface\nmarker for human myometrial stem/progenitor cells. Commun. Biol.\n6, 686 (2023).\n81. Salama, S. A., Diaz-Arrastia, C. R., Kilic, G. S. & Kamel, M. W.\n2-Methoxyestradiol causes functional repression of transforming\ngrowth factor beta3 signaling by ameliorating Smad and non-Smad\nsignaling pathways in immortalized uterine fibroid cells. Fertil. Steril.\n98, 178–184 (2012).\n82. Uimari, O., Subramaniam, K. S., Vollenhoven, B. & Tapmeier, T. T.\nUterine Fibroids (Leiomyomata) and Heavy Menstrual Bleeding.\nFront. Reprod. Health 4, 818243 (2022).\n83. Ciarmela, P. et al. Growth factors and myometrium: biological\neffects in uterine fibroid and possible clinical implications. Hum.\nReprod. Update 17, 772–790 (2011).\n84. Reschke, L. et al. Leptin induces leiomyoma cell proliferation and\nextracellular matrix deposition via JAK2/STAT3 and MAPK/ERK\npathways. F. S Sci. 3, 383–391 (2022).\n85. Geuens, T., Bouhy, D. & Timmerman, V. The hnRNP family: insights into\ntheir role in health and disease.Hum. Genet 135,8 5 1–867 (2016).\n86. Thibault, P. A. et al. hnRNP A/B Proteins: An Encyclopedic\nAssessment of Their Roles in Homeostasis and Disease. Biology\n(Basel) 10, https://doi.org/10.3390/biology10080712 (2021).\n87. Alarcon, C. R. et al. HNRNPA2B1 is a mediator of m(6)A-dependent\nnuclear RNA processing events. Cell 162, 1299–1308 (2015).\n88. Pankov, R. & Yamada, K. M. Fibronectin at a glance. J. Cell Sci. 115,\n3861–3863 (2002).\n89. Dalton, C. J. & Lemmon, C. A. Fibronectin: Molecular structure,\nfibrillar structure and mechanochemical signaling. Cells 10, https://\ndoi.org/10.3390/cells10092443 (2021).\n90. Spada, S., Tocci, A., Di Modugno, F. & Nistico, P. Fibronectin as a\nmultiregulatory molecule crucial in tumor matrisome: from structural\nand functional features to clinical practice in oncology. J. Exp. Clin.\nCancer Res. 40, 102 (2021).\n91. Dorafshan, S. et al. Periostin: biology and function in cancer. Cancer\nCell Int. 22, 315 (2022).\n92. Amendola, I. L. S., Spann, M., Segars, J. & Singh, B. The Mediator\nComplex Subunit 12 (MED-12) gene and uterine fibroids: a\nsystematic review. Reprod. Sci. 31, 291–308 (2024).\n93. He, C. et al. Frequency of MED12 mutation in relation to tumor and\npatient’s clinical characteristics: a meta-analysis. Reprod. Sci. 29,\n357–365 (2022).\n94. Islam, M. S. et al. Extracellular matrix and Hippo signaling as\ntherapeutic targets of anti fibrotic compounds for uterine fibroids.\nClin. Transl. Med. 11, e475 (2021).\n95. Islam, M. S., Ciavattini, A., Petraglia, F., Castellucci, M. & Ciarmela,\nP. Extracellular matrix in uterine leiomyoma pathogenesis: a\npotential target for future therapeutics. Hum. Reprod. Update 24,\n59–85 (2018).\n96. Yang, Q. & Al-Hendy, A. Update on the role and regulatory\nmechanism of extracellular matrix in the pathogenesis of uterine\nfibroids. Int J. Mol. Sci. 24, 5778 (2023).\n97. Stewart, E. A., Friedman, A. J., Peck, K. & Nowak, R. A. Relative\noverexpression of collagen type I and collagen type III messenger\nribonucleic acids by uterine leiomyomas during the proliferative\nphase of the menstrual cycle. J. Clin. Endocrinol. Metab. 79,\n900–906 (1994).\n98. Behera, M. A. et al. Thrombospondin-1 and thrombospondin-2\nmRNA and TSP-1 and TSP-2 protein expression in uterine\nfibroids and correlation to the genes COL1A1 and COL3A1 and to\nthe collagen cross-link hydroxyproline. Reprod. Sci. 14,6 3 –76\n(2007).\n99. Iwahashi, M. et al. Immunohistochemical analysis of collagen\nexpression in uterine leiomyomata during the menstrual cycle. Exp.\nTher. Med. 2, 287–290 (2011).\n100. Malik, M., Norian, J., McCarthy-Keith, D., Britten, J. & Catherino, W.\nH. Why leiomyomas are called fibroids: the central role of\nextracellular matrix in symptomatic women. Semin. Reprod. Med.\n28, 169–179 (2010).\n101. Leppert, P. C. et al. Comparative ultrastructure of collagen fibrils in\nuterine leiomyomas and normal myometrium. Fertil. Steril. 82,\n1182\n–1187 (2004).\n102. Bogusiewicz, M. et al. Expression of matricellular proteins in human\nuterine leiomyomas and normal myometrium. Histol. Histopathol.\n27, 1495–1502 (2012).\n103. Patel, A., Malik, M., Britten, J., Cox, J. & Catherino, W. H.\nMifepristone inhibits extracellular matrix formation in uterine\nleiomyoma. Fertil. Steril. 105, 1102–1110 (2016).\n104. Shen, X., Yang, Z., Feng, S. & Li, Y. Identi fication of uterine\nleiomyosarcoma-associated hub genes and immune cell in filtration\npattern using weighted co-expression network analysis and\nCIBERSORT algorithm. World J. Surg. Oncol. 19, 223 (2021).\n105. Courtoy, G. E. et al. Gene expression changes in uterine myomas in\nresponse to ulipristal acetate treatment.Reprod. Biomed. Online 37,\n224–233 (2018).\n106. Malik, M., Britten, J., Borahay, M., Segars, J. & Catherino, W. H.\nSimvastatin, at clinically relevant concentrations, affects human\nuterine leiomyoma growth and extracellular matrix production.Fertil.\nSteril. 110, 1398–1407.e1391 (2018).\n107. Navarro, A., Bariani, M. V., Yang, Q. & Al-Hendy, A. Understanding\nthe impact of uterinefibroids on human endometrium function.Front\nCell Dev. Biol. 9, 633180 (2021).\n108. Zhang, Y., Qian, J., Gu, C. & Yang, Y. Alternative splicing and cancer:\na systematic review. Signal Transduct. Target Ther. 6, 78 (2021).\n109. Cheng, Z., Shang, Y., Gao, S. & Zhang, T. Overexpression of\nU1 snRNA induces decrease of U1 spliceosome function associated\nwith Alzheimer’s disease. J. Neurogenet. 31, 337–343 (2017).\n110. Takayama, K. I. et al. Dysregulation of spliceosome gene expression\nin advanced prostate cancer by RNA-binding protein PSF. Proc.\nNatl. Acad. Sci. USA 114, 10461–10466 (2017).\n111. Zhao, Y. & Young, S. L. TGF-beta regulates expression of tenascin\nalternative-splicing isoforms in fetal rat lung. Am. J. Physiol. 268,\nL173–L180 (1995).\n112. Tripathi, V. & Zhang, Y. E. Redirecting RNA splicing by SMAD3 turns\nTGF-beta into a tumor promoter. Mol. Cell Oncol. 4, e1265699\n(2017).\n113. Weg-Remers, S., Ponta, H., Herrlich, P. & Konig, H. Regulation of\nalternative pre-mRNA splicing by the ERK MAP-kinase pathway.\nEMBO J. 20, 4194–4203 (2001).\n114. Pelisch, F., Blaustein, M., Kornblihtt, A. R. & Srebrow, A. Cross-talk\nbetween signaling pathways regulates alternative splicing: a novel\nrole for JNK. J. Biol. Chem. 280, 25461–25469 (2005).\n115. Goncalves, V., Matos, P. & Jordan, P. Antagonistic SR proteins\nregulate alternative splicing of tumor-related Rac1b downstream of\nthe PI3-kinase and Wnt pathways. Hum. Mol. Genet. 18, 3696–3707\n(2009).\n116. Chang, J. W. et al. mTOR-regulated U2af1 tandem exon splicing\nspecifies transcriptome features for translational control. Nucleic\nAcids Res. 47, 10373–10387 (2019).\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 15\n\n117. Lee, F. F. et al. NF-kappaB mediates lipopolysaccharide-induced\nalternative pre-mRNA splicing of MyD88 in mouse macrophages. J.\nBiol. Chem. 295, 6236–6248 (2020).\n118. Chu, W. K., Hung, L. M., Hou, C. W. & Chen, J. K. PKC regulates YAP\nexpression through alternative splicing of YAP 3’UTR Pre-mRNA by\nhnRNP F. Int J. Mol. Sci. 22, 694 (2021).\n119. Iborra, A., Mayorga, M., Llobet, N. & Martinez, P. Expression of\ncomplement regulatory proteins [membrane cofactor protein\n(CD46), decay accelerating factor (CD55), and protectin (CD59)] in\nendometrial stressed cells. Cell Immunol. 223,4 6–51 (2003).\n120. Nogawa Fonzar-Marana, R. R. et al. Expression of complement\nsystem regulatory molecules in the endometrium of normal\novulatory and hyperstimulated women correlate with menstrual\ncycle phase. Fertil. Steril. 86, 758–761 (2006).\n121. Hiroi, H. et al. Expression and regulation of periostin/OSF-2 gene in\nrat uterus and human endometrium. Endocr. J. 55, 183–189 (2008).\n122. Xu, X. et al. Periostin enhances migration, invasion, and adhesion of\nhuman endometrial stromal cells through Integrin-Linked Kinase 1/\nAkt signaling pathway. Reprod. Sci. 22, 1098–1106 (2015).\n123. Cao, W., Mah, K., Carroll, R. S., Slayden, O. D. & Brenner, R. M.\nProgesterone withdrawal up-regulates fibronectin and integrins\nduring menstruation and repair in the rhesus macaque\nendometrium. Hum. Reprod. 22, 3223–3231 (2007).\n124. Bilalis, D. A., Klentzeris, L. D. & Fleming, S. Immunohistochemical\nlocalization of extracellular matrix proteins in luteal phase\nendometrium of fertile and infertile patients. Hum. Reprod. 11,\n2713–2718 (1996).\n125. Borahay, M. A., Al-Hendy, A., Kilic, G. S. & Boehning, D. Signaling\npathways in Leiomyoma: Understanding pathobiology and\nimplications for therapy. Mol. Med. 21, 242–256 (2015).\n126. Bordeleau, F. et al. Tissue stiffness regulates serine/arginine-rich\nprotein-mediated splicing of the extra domain B-fibronectin isoform\nin tumors. Proc. Natl. Acad. Sci. USA 112, 8314–8319 (2015).\n127. Islam, M. S., Parish, M., Brennan, J. T., Winer, B. L. & Segars, J. H.\nTargeting fibrotic signaling pathways by EGCG as a therapeutic\nstrategy for uterine fibroids. Sci. Rep. 13, 8492 (2023).\n128. Perez-Riverol, Y. et al. The PRIDE database resources in 2022: a hub\nfor mass spectrometry-based proteomics evidences. Nucleic Acids\nRes. 50, D543–D552 (2022).\nAcknowledgements\nThis work was supported through the Bayer - Oxford Alliance in Women’s\nHealthcare, which receives funding through the NIHR Biomedical Research\nCentre, the Endometriosis CaRe Centre Oxford, Oxford University Medical\nSciences Division and Bayer Healthcare. Further research support was\nobtained from Innovate UK (UO, MP, APC), the National Institute for Health\nResearch Oxford Biomedical Research Centre (UO), Cancer Research UK\n(CRUK, UO), the Bone Cancer Research Trust (APC and UO), the Leducq\nEpigenetics of Atherosclerosis Network (LEAN) programme grant from the\nLeducq Foundation (UO), the Chan Zuckerberg Initiative (APC) and the\nMyeloma Single Cell Consortium (UO). APC is a recipient of an MRC Career\nDevelopment Fellowship (MR/V010182/1). Work in the BMK laboratory was\nsupported by the Wellcome Trust (097812/Z/11/Z) and the Engineering and\nPhysical Science Research Council (EP/N034295/1).\nAuthor contributions\nU.O., CY.W., M.P. and A.P.C. designed and supervised the study; C.Y.W.,\nM.P. and U.O. wrote the first manuscript draft. CY.W., A.P.C. and U.O.\nrevised the draft versions of the manuscript. K.Z., C.M.B., J.M. (Oxford),\nK.G., S.M., M.M. supervised and performed sample collection and clinical\nannotation, with important help from C.M.B., T.M.Z. and A.L.H., C.Y.W.,\nM.P., D.O.B., J.M. (Oxford), N.M., V.G., B.M., S.B., R.F. performed\nexperiments. C.Y.W., D.O.B., A.P.C., J.M. (Bayer) performed data analysis,\nwith significant contributions from A.N., M.O., B.K., and A.L.H. C.M.B., K.Z.,\nA.L.H., S.M., J.M. (Bayer), N.S. and T.M.Z. contributed critical data\ninterpretation. All authors have read and provided input to the manuscript.\nCompeting interests\nFS, MO, NS, JM, and TMZ are employees and shareholders of Bayer\nPharmaceuticals. MP, APC and UO are co-founders of Caeruleus Genomics\nplc. The study was jointly supported by Oxford and Bayer Healthcare;\nconceptualisation, research, data analysis and presentation were con-\nducted in an unbiased manner and not influenced by the funding bodies.\nAdditional information\nSupplementary informationThe online version contains\nsupplementary material available at\nhttps://doi.org/10.1038/s43856-025-01051-x\n.\nCorrespondenceand requests for materials should be addressed to\nUdo Oppermann or Adam P. Cribbs.\nPeer review information Communications Medicinethanks Md Sorifol\nIslam and the other, anonymous, reviewer(s) for their contribution to the peer\nreview of this work. [A peer review file is available.]\nReprints and permissions informationis available at\nhttp://www.nature.com/reprints\nPublisher’s note Springer Nature remains neutral with regard to\njurisdictional claims in published maps and institutional affiliations.\nOpen Access This article is licensed under a Creative Commons\nAttribution 4.0 International License, which permits use, sharing,\nadaptation, distribution and reproduction in any medium or format, as long\nas you give appropriate credit to the original author(s) and the source,\nprovide a link to the Creative Commons licence, and indicate if changes\nwere made. The images or other third party material in this article are\nincluded in the article ’s Creative Commons licence, unless indicated\notherwise in a credit line to the material. If material is not included in the\narticle’s Creative Commons licence and your intended use is not permitted\nby statutory regulation or exceeds the permitted use, you will need to\nobtain permission directly from the copyright holder. To view a copy of this\nlicence, visit http://creativecommons.org/licenses/by/4.0/.\n© The Author(s) 2025\n1Botnar Research Centre, NIHR BRC, University of Oxford, Oxford, UK.2Target Discovery Institute, Centre for Medicines Discovery, Nuffield Department of Medicine,\nUniversity of Oxford, Oxford, UK.3Nuffield Department of Women’s & Reproductive Health, University of Oxford, Oxford, UK.4Department of Oncology, University of\nOxford, Oxford, UK.5Research and Early Development, Bayer AG, Berlin, Germany.6These authors contributed equally: Chen-Yi Wang, Martin Philpott.\ne-mail: Udo.oppermann@ndorms.ox.ac.uk; Adam.cribbs@ndorms.ox.ac.uk\nhttps://doi.org/10.1038/s43856-025-01051-x Article\nCommunications Medicine |           (2025) 5:318 16","source_license":"CC0","license_restricted":false}