henhaobio
Sign in

The FASEB JournalPhysiologyPeer-reviewed journal article

Dynamic Transcriptome, Accessible Genome and PGR Cistrome Profiles in the Human Myometrium

San‐Pin WuiD, Matthew Linley AndersoniD, Tianyuan WangiD, Lecong ZhouiD, Olivia M. EmeryiD, Xilong LiiD, Francesco John DeMayoiD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

The myometrium undergoes structural and functional remodeling during pregnancy. We hypothesize that myometrial genomic elements alter correspondingly in preparation for parturition. Human myometrial tissues from nonpregnant (NP) and term pregnant (TP) human subjects were examined by RNAseq, ATACseq and PGR ChIPseq assays to profile transcriptome, assessible genome and PGR occupancy. NP and TP specimens exhibit 2890 differentially expressed genes, reflecting an increase of metabolic, inflammatory and PDGF signaling, among others, in adaptation to pregnancy. At the epigenome level, patterns of accessible genome change between NP and TP myometrium, leading to altered enrichment of binding motifs for hormone and muscle regulators such as the progesterone receptor (PGR), Kruppel-like factors and MEF2A transcription factors. PGR genome occupancy exhibits a significant difference between the two stages of the myometrium, concomitant with distinct transcriptomic profiles including genes such as ENO1, LHDA, and PLCL1 in the glycolytic and calcium signaling pathways. Over-representation of SRF, MYOD and STAT binding motifs in PGR occupying sites further suggests interactions between PGR and major muscle regulators for myometrial gene expression. In conclusion, changes in accessible genome and PGR occupancy are part of the myometrial remodeling process and may serve as mechanisms to formulate the state-specific transcriptome profiles.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (27 Nov 2019) · CC0 1.0

Term-pregnant versus nonpregnant human myometrium shows 2890 differentially expressed genes, more open chromatin regions (85777 vs 45501), and progesterone receptor binding shifted from promoters toward distal sites.

Myometrial disorders can lead to preterm birth and dystocia, and progesterone signaling through the progesterone receptor (PGR) holds back inflammatory and contractile activity before parturition. Earlier genome-wide studies looked for mechanisms of the switch from term nonlabor to labor; this work instead profiles how human myometrium remodels from the nonpregnant to the term-pregnant state, measuring transcripts, open chromatin and PGR binding.

Key results

  1. ShownPregnancy mainly changes how much each myometrial gene is expressed, not which genes are expressed. 2890 differentially expressed genes (DEGs) separate nonpregnant (NP) from term-pregnant (TP) tissue, and genes linked to parturition rise toward term in two independent cohorts.

    RNA-seq, 3 NP vs 3 TP: 15907 active genes (FPKM ≥ 1), 14612 shared (95.7% of NP and 93.4% of TP expressed genes); principal component analysis separates stages; 2890 DEGs (|fold change| ≥ 1.4, adjusted p < 0.05), 1043 higher in NP and 1847 in TP. OXTR, GJA1, ZEB1, PLCL1 mean fold changes 156, 2.9, 1.6, 2.6; increases also seen by qRT-PCR in separate specimens (6 NP, 9 TP). · Adaptive Myometrial Transcriptome for Pregnancy; RNAseq; Quantitative RT-PCR

  2. ShownGlycolytic genes are expressed more highly at term. The authors read this as greater glucose use and more capacity to produce lactate, which they propose may help keep the myometrium quiescent.

    SLC2A1, ALDOB, PGK1, ENO1, PKM2 and LDHA mRNA higher at TP in RNA-seq (3 vs 3) and a separate qRT-PCR cohort (6 NP vs 9 TP, two-tailed t-test); HK2 significant in one cohort, only a trend in the other. Ingenuity Pathway Analysis (IPA) predicts a glucose-response pattern consistent with greater glucose use at TP, plus increased HIF1A activity that the authors read as a hypoxic myometrium at term. · Adaptive Myometrial Transcriptome for Pregnancy; An Indication of Progesterone Receptor on…; Quantitative RT-PCR

  3. SuggestedTerm myometrium has more open chromatin regions (OCRs). MEF2A motifs are enriched in TP-only regions, and estrogen receptor and E2F1 motifs in NP-only regions. Stage-specific OCRs near DEGs change either with or against the expression of those genes.

    ATAC-seq, 3 NP vs 3 TP: 45501 vs 85777 OCRs; 9564 NP-only, 35937 shared, 50840 TP-only, with 128, 82 and 121 HOMER-enriched motifs respectively. Within 25 kb of DEG transcription start sites, 2989 OCRs change in the same direction as 1461 DEGs and 1660 OCRs opposite to 980 DEGs. BaGFoot footprint motifs differ by stage (e.g., SP and EGR1 in NP; FOX and HOX in TP). · Accessible Genome Dynamics in the Myometrium; ATACseq; Novelty of the study

  4. SuggestedPGR binding shifts between stages. Nonpregnant peaks sit more at promoters, and NP-distinct sites favor ETS, NF-Y, CTCF, YY1, Sp1 and NRF1 motifs. TP-distinct sites favor PGR/glucocorticoid receptor, AP-1 and TEAD motifs.

    PGR ChIP-seq, 2 specimens per stage at 9 million reads: 24095 and 9778 peaks (NP), 10717 and 8053 (TP); promoter share 34.3% and 35.4% (NP) vs 26.4% and 12.0% (TP); PGR consensus motif in 5.01% and 18.57% (NP) vs 21.33% and 26.20% (TP) of peaks. Of stage-common sites, 2220 are shared, 2773 NP-distinct and 1047 TP-distinct; YY1 motif p 10^-55 and 10^-61 in NP vs 10^-20 and > 0.01 in TP. · PGR Genome Occupancy in the Myometrial Tissues; ChIPseq; Figure legends

  5. SuggestedAssigning each PGR site to its nearest gene gives 2203 expressed genes as candidate direct PGR targets. ChIP-qPCR in separate specimens finds less PGR at ENO1 and LDHA at term, even though their mRNA rises, and confirms PGR at a site 93 kb upstream of PLCL1.

    Of 6040 shared or distinct PGR regions, 4355 (72%) map to 2203 active genes (1431 NP-distinct, 183 TP-distinct, 589 shared), with 51, 37 and 166 IPA pathways at p < 0.05 respectively and none significant in all three. ChIP-qPCR in 2 NP and 2 TP independent specimens used technical triplicates with Untr12 (gene desert) as negative control. · Candidate PGR Direct Downstream Target Genes in…; ChIP-qPCR; Tables

Methods, data and reproducibility

Bulk RNA-seq (3 NP, 3 TP), ATAC-seq (3 NP, 3 TP) and PGR ChIP-seq (2 NP, 2 TP) on surgical myometrial biopsies from premenopausal nonpregnant women and non-laboring women at term cesarean section. The three data types were integrated using nearest-gene and motif-enrichment analyses. Selected genes and PGR sites were checked by qRT-PCR (6 NP, 9 TP) and ChIP-qPCR (2 NP, 2 TP) in separate specimens.

  • Myometrial biopsies for RNA-seq: nonpregnant premenopausal (proliferative phase) vs non-laboring term pregnant (38-42 weeks, cesarean) · human · n = 3 NP and 3 TP · GSE137552 (GEO SuperSeries)
  • Myometrial biopsies for ATAC-seq, NP vs TP · human · n = 3 NP and 3 TP · GSE137552 (GEO SuperSeries)
  • Myometrial biopsies for PGR ChIP-seq, NP vs TP, with myometrium input as control · human · n = 2 NP and 2 TP · GSE137552 (GEO SuperSeries)
  • Independent myometrial specimens for qRT-PCR of parturition-associated and glycolytic genes · human · n = 6 NP and 9 TP
  • Independent myometrial specimens for PGR ChIP-qPCR at ENO1, LDHA and PLCL1 loci · human · n = 2 NP and 2 TP
  • Public mouse myometrium transcriptome (microarray), 18.5 days post coitus vs virgin, used for IPA comparison · mouse · GSE17021
Methods & tools
TopHat 2.0.4, Cufflinks 2.2.1, Bowtie 1.1.2, MACS2 (2.1.1.20160309), HOMER, BaGFoot, Picard MarkDuplicates (picard-tools-1.119), BEDTools 2.27.1, Ingenuity Pathway Analysis (IPA), DAVID v6.8, Cis-regulatory Element Annotation System, anti-PGR antibody sc-7208 (Santa Cruz)
Reagents & environment
Versions stated: TopHat 2.0.4, Cufflinks 2.2.1, Bowtie 1.1.2, MACS 2.1.1.20160309, picard-tools-1.119, BEDTools 2.27.1, DAVID v6.8; reads aligned to hg19. No container or requirements file.
Reproducibility gaps
No analysis code released; ChIP-seq downsampling used custom scripts that are not provided. The raw data accession (GSE137552) is given without a URL.
Limitations (6)
  • Bulk tissue profiles mix smooth muscle with blood and vascular cells, which can dilute the signal from the cell type of interest; the authors note single-cell assays would give finer resolution. · stated by the authors
  • Sample sizes for RNA-seq, ATAC-seq and ChIP-seq are small, limited by biopsy size and by the difficulty of chromatin immunoprecipitation and ATAC library preparation on muscle tissue. · stated by the authors
  • The PGR antibody recognizes both PGR-A and PGR-B, so isoform-specific binding and the stage-dependent ratio of the two isoforms cannot be resolved. · stated by the authors
  • Each PGR site is linked only to its nearest gene, so long-range enhancer-gene contacts are missed without chromatin conformation data. · stated by the authors
  • Stage is confounded with serum progesterone (low in proliferative-phase nonpregnant women, high at term), and only the two ends of the pregnancy spectrum were sampled. · stated by the authors
  • The enhancer and repressor roles assigned to open regions, and the PGR target calls, rest on correlation and genomic proximity with no perturbation test. PGR peak numbers also vary widely within a stage (24095 vs 9778 in the two NP specimens). · noted in this summary

Read this if you study uterine smooth muscle, progesterone receptor genomics or preterm birth and want human myometrial RNA-seq, ATAC-seq and PGR ChIP-seq profiles comparing nonpregnant with term-pregnant tissue.

Citations
28
FWCI
0.58
Field percentile
top 37%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISWu '24Khader '26Khader '25Chi '19Shchuka '20Mbadhi '26
78% semantic matchCell Biology
S. P. S. Wu, E. Quiroz, T. Wang+5 more·bioRxiv·Feb 19, 2024