Abstract

Background

   A balance between estrogen (E2) and progesterone (P4) is vital for a
   successful pregnancy, and an imbalance between these two hormones
   yields female infertility. E2 mediates uterine receptivity and
   regulates endometrial growth, the immune system, and hormone signaling
   by rapidly inducing early growth response 1 (EGR1). However, the
   precise mechanism by which EGR1 regulates E2-mediated uterine growth
   remains unclear. This study examined the transcriptional signatures of
   ovariectomized (OVX) Egr1 knockout (KO) mice compared to those of
   wild-type (WT) mice to clarify the function of EGR1 in the E2-dependent
   uterine response.

Results

   Egr1 KO uteri exhibited an impaired E2 response, with significant
   changes in the expression of the key genes Bgn, c-Kit, Ripor2, and
   Arg2. During early E2 response, Egr1 KO uteri showed upregulated
   insulin-like growth factor (IGF) signaling pathway genes and
   downregulated reproduction-related genes. During late E2 response, Egr1
   KO uteri showed enhanced proliferative processes, such as DNA
   replication and mitotic cell cycle phase transition, potentially
   related to poly-ADP ribosylation (PARylation), along with a reduction
   in immune response. Notably, the transcriptional signatures in mature
   OVX Egr1 KO uteri resembled those in immature uteri, with similar
   increases in proliferation and decreases in immune response at the
   pathway level.

Conclusions

   Our findings indicate that EGR1 is essential for regulating immune
   response and uterine proliferation via IGF signaling and PARylation,
   and acts as a gatekeeper transcription factor that mediates E2 actions
   in the mature uterus. Notably, we found that the transcriptional
   signatures of EGR1 in mature uteri overlapped with the primary E2
   function and suggested a novel concept that these transcriptional
   signatures in mature Egr1 KO uteri are similar to those of immature
   uteri. Our findings offer novel insights into the role of EGR1 as an E2
   mediator in the uterus at the molecular level.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12864-025-11904-3.

   Keywords: Early growth response 1, Estrogen, Insulin-like growth
   factor, Proliferation, Immune response, Ovariectomized uterus, Immature
   uterus

Background

   In the uterine reproductive cycle, maintaining a proper balance between
   estrogen (E2) and progesterone (P4) is important for establishing and
   sustaining a successful pregnancy [[28]1–[29]3]. This cycle consists of
   two phases: the E2-dominated proliferative phase and the secretory
   phase, which is influenced by P4 and E2 [[30]4]. In the proliferative
   phase, the E2 response is further divided into early and late stages
   [[31]5, [32]6]. In the early stage, E2 induces water imbibition,
   leading to an increase in uterine weight; meanwhile, in the late stage,
   E2 promotes epithelial cell proliferation through the estrogen
   receptor-alpha (ERα)-mediated insulin-like growth factor (IGF)
   signaling pathway [[33]7, [34]8]. This E2 response is fundamental to
   uterine growth and implantation receptivity, whereas, during the
   secretory phase, P4 modulates E2 activity to induce uterine maturation
   [[35]9]. Disruption of this hormonal balance, resulting from impaired
   P4 signaling and uncontrolled E2 activity, contributes to reproductive
   disorders and various diseases in females, including endometriosis and
   endometrial cancer [[36]10]. Therefore, it is necessary to understand
   the genes regulated by ovarian hormones, such as P4 or E2, during the
   uterine reproductive cycle to provide effective treatments [[37]11].

   Notably, early growth response 1 (Egr1) is induced by E2 for
   fine-tuning uterine receptivity [[38]12]. Egr1, a member of the Egr
   family, is a zinc-finger transcription factor involved in regulating
   cell proliferation, differentiation, and apoptosis [[39]13, [40]14] and
   can be induced by various extracellular signals, including growth
   factors, cytokines, hypoxia, and physical stimuli. EGR1 is known to
   respond early and influence the long-term effects of these signals
   [[41]15, [42]16]. Likewise, E2 rapidly and transiently induces Egr1 in
   uterine epithelial cells and stroma via the nuclear ERα-ERK1/2 pathway
   [[43]17]. Previous studies have reported that EGR1 participates in the
   transcriptional regulation of E2-induced genes in the mouse uterus
   [[44]18, [45]19] and balances E2 and P4 during the peri-implantation
   period [[46]12, [47]20]. The absence of EGR1 in the mouse uterus leads
   to a lack of progesterone receptor expression in uterine epithelial
   cells, impairing P4 signaling [[48]12]. This leads to the
   hyperproliferation of epithelial cells, inadequate differentiation, and
   failure to establish proper uterine receptivity. Taken together, EGR1
   plays a crucial role in balancing ovarian hormone signaling and
   regulating E2-regulated genes for successful implantation.

   Despite advances in understanding the importance of EGR1 in female
   reproduction, the underlying molecular mechanisms and regulatory
   properties are unknown. In particular, the function of EGR1 during the
   E2-dominated phase of uterine growth and development is unclear. This
   is critical because it is key to understanding how E2-driven gene
   regulation is fine-tuned to ensure proper uterine receptivity.

   Studies on E2 signaling in the uterus have commonly employed mature
   ovariectomized (OVX) and immature uterus models in uterotrophic assays
   to evaluate the uterine response to exogenously administered E2
   [[49]21]. In the mature OVX mouse model, animals are raised for 6 to 8
   weeks after birth, at which time they reach sexual maturity and then
   undergo OVX. To ensure that the uterus regresses to a stable baseline
   level, an adequate period of at least 14 days is required before the
   first exogenous E2 administration. In contrast, the immature mouse
   model refers to animals in which the hypothalamic-pituitary-gonadal
   axis is not yet fully activated, and endogenous E2 levels have not
   begun to rise, indicating that uterine maturation induced by endogenous
   E2 does not occur. These models are suitable for studying E2 responses
   since both respond sensitively to exogenous E2. However, differences in
   uterine maturation lead to a distinct E2 response in each of the two
   models [[50]22], and the complex mechanisms underlying these
   differences have not yet been fully elucidated.

   In this study, we investigated the role of EGR1 in the E2-dominant
   phase by comparing uterine transcriptional patterns in mature OVX Egr1
   knockout (KO) and wild-type (WT) mice using time-series microarrays. We
   further analyzed time-series microarray data from immature uteri
   treated with E2 to ultimately assess the impact of EGR1 on the E2
   response across different uterine maturation states. This dual approach
   not only clarifies the specific function of EGR1 in an E2-dominant
   context but also provides novel insights into how EGR1 influences the
   differences in E2 responses between mature OVX and immature uterine
   models.

Methods

Microarray data for Egr1 knockout, wild-type, and immature mice

   In one of our previous studies, we generated time-series microarrays to
   compare the E2 responses between Egr1 KO and WT mice [[51]12]. We
   reanalyzed these microarray data using advanced bioinformatic methods
   to elucidate EGR1-mediated transcriptional regulation during the
   E2-dominated phase. All animal procedures were approved by the
   Institutional Animal Care and Use Committee of CHA University (140030).
   Egr1 KO (Egr1^−/−) mice, aged 6 to 8 weeks, were generously provided by
   Dr. Jeffrey Milbrandt (Washington University, St. Louis, MO, USA), and
   WT (Egr1^+/+) mice were used as controls. All mice underwent OVX prior
   to E2 treatment. Microarrays were conducted using total uterine RNA
   collected 0, 3, 6, and 24 h after E2 treatment. The Illumina Mouse WG-6
   V2 (48 K) array (Illumina, San Diego, CA, USA) was hybridized with
   complementary RNA probes at ISTECH (Goyang, South Korea).

   To compare the E2-induced transcriptome in Egr1 KO uteri with that of
   immature (IM) uteri, we downloaded a microarray dataset ([52]GSE2195)
   [[53]23] from the GEO database. This dataset includes samples from
   female Alpk: Ap[f]CD-1 mice aged 19 to 20 days. In this dataset, the
   treatment group received a single subcutaneous injection of E2, while
   the control group was injected with arachis oil, and samples were
   collected at 1, 2, 4, 8, and 24 h post-treatment. We used the ComBat
   function of the R package sva (version 3.48.0) to correct for batch
   effects between the datasets of IM and KO mice.

Microarray data analysis and pathway enrichment test

   Microarray data were quantified using the BeadStudio method and
   processed with quantile normalization. To identify differentially
   expressed genes (DEGs) at each time point, we used the R package limma
   (version 3.56.1) [[54]24]. To address the multiple comparison problem,
   we adjusted p-values using the Benjamini-Hochberg method. DEGs were
   determined with an adjusted p-value < 0.05 and a fold change > 1.5 or <
   −1.5 as the criteria. For pathway enrichment analysis of DEGs, the
   enrichGO function of the R package clusterProfiler (version 4.8.1)
   [[55]25] was used based on the Gene Ontology (GO) database.
   Significantly enriched GO pathways in the DEGs were identified with an
   adjusted p-value < 0.05.

Protein-protein interaction network analysis

   To identify DEG clusters, we combined protein-protein interaction (PPI)
   networks and DEGs using Cytoscape (version 3.10.1) [[56]26]. In
   Cytoscape, we downloaded PPI networks of DEGs from STRING (protein full
   network of Mus musculus) using stringApp (v2.0.1) and clustered them
   using the MCL clustering method of clusterMaker2 (v2.3.4). For the
   functional analysis of each MCL cluster containing five or more DEGs,
   we performed pathway enrichment tests in each cluster using default
   values.

Inferring PARylation and glycosylation activity score

   To measure poly-ADP ribosylation (PARylation) and glycosylation
   activity, we calculated gene set signature scores by gene set variation
   analysis (GSVA) using the R package GSVA (version 1.48.0) [[57]27].
   Gene lists for PARylation and glycosylation target proteins were
   obtained from previous studies [[58]28, [59]29]. Since these genes are
   listed by human gene symbols, we converted them to mouse EntrezGene IDs
   using MGI Human and Mouse Homology Data (data date:11/03/2023; version
   MGI 6.22) [[60]30]. We used limma (version 3.56.1) to statistically
   measure the differences in signature scores.

Cell-type enrichment test

   The deconvolute function of the immunedeconv R package (version 2.1.0)
   was used to calculate immune and stromal scores [[61]31]. The xCell
   algorithm was used as the deconvolution function.

Identifying key hallmark gene sets and genes in Egr1 KO and immature uteri

   We calculated GSVA scores of the hallmark gene sets in the mSigDB
   database [[62]32] and analyzed them using principal component analysis
   (PCA) to assess the similarity between the Egr1 dataset and the IM mice
   dataset. Gene sets with contributions higher than the average in a
   principal component (PC) were considered significant for interpreting
   that PC [[63]33]. We performed three steps to identify key genes of the
   gene sets contributing to the similarity between Egr1 KO and IM uteri:
   first, we conducted one-way analysis of variance (ANOVA) for the
   following three groups: Egr1 WT, Egr1 KO, and IM uteri, and then
   adjusted the ANOVA p-values using the Benjamini-Hochberg method.
   Second, we applied the Tukey honest significant difference test to
   genes with an adjusted p-value < 0.05. Finally, we determined key genes
   based on the following criteria: (i) adjusted p-values < 0.05 in both
   Egr1 KO vs. Egr1 WT and IM vs. Egr1 WT comparisons and (ii) no
   significant differences (adjusted p-value > 0.05) in Egr1 KO vs. IM
   comparison.

Statistical analysis

   All values represent means ± standard deviations (SD). Heatmaps,
   volcano plots, bar plots, dot plots, box plots, and line graphs were
   generated using R (version 4.4.0) [[64]34]. All analyses were performed
   in R, except for the protein-protein interaction network analysis,
   which was conducted using Cytoscape (version 3.10.1).

Results

Dysregulation of the E2 response in Egr1 KO uteri was the highest 24 h after
E2 treatment

   We performed temporal profiling of Egr1 KO and WT uteri at 3, 6, and
   24 h after E2 treatment to investigate the function of EGR1 in the
   uterus (Fig. [65]1A). Since EGR1 mRNA and protein levels peaked 2 h
   after E2 treatment [[66]17], we defined 3 h as the early E2 response,
   6 h as the intermediate E2 response, and 24 h as the late E2 response
   in this study. In profiling, all samples were grouped according to the
   elapsed time after E2 treatment (Fig. [67]1B), regardless of Egr1
   status. Because E2 has a greater effect on the uterus than EGR1, it is
   reasonable that E2-induced transcript expression changes are stronger
   than those associated with EGR1 deficiency. Moreover, the gene
   expression profile in the intermediate E2 response (6 h) clustered with
   that in the early E2 response (3 h) owing to similar expression
   pattern. Gene expression in the late E2 response (24 h) clustered with
   the pre-treatment time point (0 h), as some genes returned to their
   baseline state in the late E2 response, despite being altered in the
   early or intermediate E2 response.

Fig. 1.

   [68]Fig. 1
   [69]Open in a new tab

   Study design and results of mRNA expression analysis. A Design of the
   mRNA expression analysis. B Heatmap of hierarchical clustering based on
   gene expression; genes with the highest IQR (top 1000) in all samples
   were used for clustering. C Volcano plot depicting DEGs with consistent
   expression changes and bar plot showing the number of DEGs at each time
   point. The X-axis represents the log2 fold change in gene expression
   levels in Egr1 KO uteri compared to those in WT uteri, and the Y-axis
   represents the − log10 adjusted p-value; gray dots represent
   non-significant genes. Upregulated DEGs are depicted as red dots and
   downregulated DEGs are depicted as blue dots. IQR, interquartile range;
   DEG, differentially expressed gene; E2, estrogen

   Based on the DEG analysis comparing Egr1 KO with WT uteri at 3, 6, and
   24 h (Fig. [70]1C and Supplementary Table 1), we identified 176, 80,
   and 303 DEGs, respectively. The number of DEGs 24 h after E2 treatment
   was the highest, with 202 downregulated and 101 upregulated DEGs in
   Egr1 KO compared to WT mice. The lowest number of DEGs was 80 at 6 h,
   of which 37 were downregulated and 43 were upregulated. The change in
   the transcriptional profile in E2-treated Egr1 KO uteri was the largest
   in the late response and smallest in the intermediate response. Some of
   these DEGs, such as Bgn, c-Kit, Ripor2, and Arg2, showed consistent
   upregulation or downregulation in response to E2.

EGR1 regulates the IGF signaling pathway and reproduction in the early and
intermediate E2 responses

   To understand the role of EGR1 in the early and intermediate E2
   responses, we investigated DEGs at 3 and 6 h (Fig. [71]2). In DEGs
   upregulated at 3 h in Egr1 KO uteri, cell proliferation pathways were
   enriched, such as “positive regulation of mitotic nuclear division” and
   “IGF receptor signaling pathway” (Fig. [72]2A). “Killing of cells of
   another organism” was enriched in upregulated DEGs at both time
   intervals. In downregulated DEGs, “positive regulation of reproductive
   process” and “sex differentiation” were enriched at 3 h, while
   “negative regulation of cell activation” was enriched at 6 h
   (Fig. [73]2B).

Fig. 2.

   [74]Fig. 2
   [75]Open in a new tab

   Regulatory role of EGR1 in early and intermediate E2 responses in mouse
   uteri. A GO term analysis of upregulated DEGs in Egr1 KO uteri. B GO
   term analysis of downregulated DEGs in Egr1 KO uteri. C IGF cluster
   within PPI network of upregulated DEGs in Egr1 KO uteri. Table shows
   enriched biological pathways. The node border color indicates
   biological pathways. Heatmap of log2 fold change (FC) in gene
   expression levels in Egr1 KO uteri compared to those in WT uteri for
   genes in the cluster. Genes that are not DEGs at a given time point are
   represented with a log2FC of 0. D Reproductive cluster within PPI
   network of downregulated DEGs in Egr1 KO uteri. GO, gene ontology; DEG,
   differentially expressed genes; KO, knockout; FDR, false discovery
   rate; E2, estrogen; IGF, insulin-like growth factor; PPI,
   protein-protein interaction

   We conducted a cluster analysis of PPI networks using DEGs at 3 and 6 h
   to enhance our understanding of the role of EGR1 in the early and
   intermediate E2 responses (Supplementary Fig. 1, Supplementary Tables 2
   and 3). In subnetworks constructed with upregulated DEGs in Egr1 KO
   uteri, we identified a cluster associated with IGF (Fig. [76]2C). IGF1
   and IGF1R have been reported to play significant roles in mediating
   E2-induced cell proliferation in uterine epithelium through ERα [[77]7,
   [78]8].

   In subnetworks constructed with downregulated DEGs in Egr1 KO uteri, we
   identified a cluster related to reproduction (Fig. [79]2D). Given that
   the Egr1 KO uterus exhibits implantation failure and impaired
   decidualization [[80]12], this cluster provides a foundation for
   further investigating the role of EGR1 in reproductive functions.
   Adamts-1 and c-Kit in this cluster have been reported to play crucial
   roles in uterine receptivity and successful embryo implantation and to
   be transcriptionally regulated by E2-induced EGR1 [[81]35, [82]36].
   Collectively, these results suggest that EGR1 is involved in IGF
   signaling and reproductive processes in the early and intermediate E2
   responses.

Uncontrolled cell proliferation and decreased immune response in the late E2
response in Egr1 KO uteri

   We performed pathway enrichment analysis of DEGs at 24 h to explore the
   characteristics of Egr1 KO uteri in the late E2 response (Fig. [83]3A).
   Pathways associated with cell proliferation, such as “DNA replication,”
   “DNA recombination,” “mitotic cell cycle phase transition,” and
   “nuclear division,” were upregulated in Egr1 KO uteri. These results
   are consistent with our previous findings of Egr1 KO uterine epithelial
   cell hyperproliferation [[84]12]. In addition, we identified
   downregulated immune response pathways in Egr1 KO uteri, including
   “cell killing,” “leukocyte mediated immunity,” and “regulation of
   immune effector process.” These results suggest that biological
   processes related to uterine growth and immunity are dysregulated in
   Egr1 KO uteri.

Fig. 3.

   [85]Fig. 3
   [86]Open in a new tab

   Regulatory role of EGR1 in the late E2 response in mouse uteri. A GO
   term analysis results of DEGs in Egr1 KO uteri compared to WT at 24 h.
   B Parp1 expression in Egr1 KO and WT uteri at 24 h. C Gene set
   signature scores for PARylation target proteins. D Gene set signature
   scores for glycosylation target proteins. E Immune scores from xCell.
   F Stroma scores from xCell. GO, gene ontology; DEG, differentially
   expressed genes; KO, knockout; WT, wild-type; E2, estrogen; PARylation,
   poly-ADP ribosylation; ns, not significant. *P < 0.05, **P < 0.01,
   ***P < 0.001

   Furthermore, of the upregulated pathways in Egr1 KO uteri, DNA
   replication and the cell cycle are known to be influenced by
   PARylation, a post-translational modification mediated by the poly-ADP
   ribose polymerase (PARP) family [[87]28, [88]37]; Parp1 was upregulated
   in Egr1 KO uteri at 24 h (Fig. [89]3B). Therefore, we hypothesized that
   EGR1-deficient uteri are influenced by PARP1 activation and PARylation.
   To infer the PARylation activity over time after E2 treatment, we
   calculated the PARylation score in each sample as a gene set signature
   score for the target proteins of PARylation (Fig. [90]3C). In Egr1 WT
   uteri, the PARylation score increased at 3 h and remained high until
   6 h, but returned to the E2 pre-treatment score (0 h) at 24 h. However,
   in Egr1 KO uteri, the PARylation score increased at 3 h but did not
   return to the basal E2 pre-treatment score (0 h) at 24 h. Glycosylation
   scores showed no significant differences between Egr1 KO and WT uteri
   (Fig. [91]3D). These findings suggest that PARylation mediated by PARP1
   influences the dysregulated biological processes in Egr1 KO uteri
   during the late E2 response.

   Next, we measured cell type enrichment scores, which reflect immune or
   stromal cell enrichment, to determine whether the downregulated immune
   response in Egr1 KO uteri was associated with immune cell enrichment
   (Fig. [92]3E). Although not statistically significant, the immune score
   was lower in Egr1 KO than in WT mice at 24 h. There were no significant
   differences in the stromal scores between the two groups (Fig. [93]3F).

Dysregulated pathways in Egr1 KO uteri in the late E2 response were similar
to those in immature uteri

   Although both the mature OVX and IM uterus models have widely been used
   to study E2 responses in uterotrophic assays, they exhibit distinct
   molecular mechanisms [[94]22]. To examine the effect of EGR1 on the
   differences in E2 responses between those models, we analyzed
   E2-treated transcriptomic data from mature OVX Egr1 KO and WT uteri
   alongside those from IM uteri [[95]23]. Our analysis followed a
   three-step approach. First, we conducted unsupervised clustering
   analysis (Supplementary Fig. [96]2), which successfully grouped all
   samples from both datasets according to the elapsed time after E2
   treatment, regardless of Egr1 knockout or immature status. Second, we
   calculated pathway signature scores based on hallmark gene sets
   (Supplementary Fig. [97]3) and performed PCA using these signature
   scores at 24 h (the late E2 response) to measure transcriptomic
   similarities among the three groups (Egr1 KO, Egr1 WT, and IM uteri).
   Egr1 KO uteri showed remarkable similarity to IM uteri along PC1
   (Fig. [98]4A), despite the presence of Egr1 expression in IM uteri
   (Supplementary Fig. [99]4). The hallmark gene sets contributing to PC1
   were predominantly related to proliferation and immune response, such
   as Myc targets, E2f targets, and interferon response (Fig. [100]4B).
   These results suggest that the late E2 response in mature Egr1 KO uteri
   is similar to that in IM uteri, particularly in terms of uncontrolled
   cell proliferation and decreased immune response.

Fig. 4.

   [101]Fig. 4
   [102]Open in a new tab

   Similarity between Egr1 KO and IM uteri in the late E2 response.
   A Principal component analysis of Egr1 KO, WT, and IM uteri in the late
   E2 response based on hallmark gene set signature scores from GSVA.
   B Heatmap of hallmark gene set signature scores contributing to the
   first principal component (PC1). C Expression patterns of key gene sets
   contributing to PC1. Proliferation-related gene sets are shown on the
   left and immune-related gene sets on the right. D Genes from Fig. 4C
   that are commonly up- or downregulated in Egr1 KO and IM uteri compared
   to WT, identified using ANOVA. IM, immature mice; KO, knockout; WT,
   wild-type; E2, estrogen; ANOVA, analysis of variance; GSVA, gene set
   variation analysis

   Finally, we determined whether the expression of key genes involved in
   proliferation and immune response was similar in Egr1 KO and IM uteri
   (Fig. [103]4C). In the unsupervised clustering analysis, Egr1 KO and IM
   uteri clustered together (Supplementary Fig. 5). However, the
   expression of most genes differed between Egr1 KO and IM uteri, and
   only 16 genes showed similar expression in both, of which 13 were
   involved in cell proliferation, and 3 in immune response. They showed
   significantly higher or lower expression (adjusted p < 0.05) in both
   Egr1 KO and IM uteri than in WT uteri (Fig. [104]4D). Collectively, our
   findings confirmed that the late E2 response in mature OVX uteri
   lacking EGR1 closely resembles that of immature uteri.

Discussion

   We investigated the function of EGR1 during the E2-dominated
   proliferative phase in E2-treated uteri, analyzing the transcriptome
   data of Egr1 KO and WT uteri 3, 6, and 24 h post-E2 treatment
   (Fig. [105]1A). While we examined EGR1 function in both E2 and P4
   contexts in our previous study [[106]12], this study aimed to isolate
   and characterize the specific role of EGR1 in E2 signaling without P4
   intervention owing to the intricate crosstalk between the E2 and P4
   signaling pathways. Furthermore, we compared our transcriptome data
   with those from immature uteri to examine the impact of EGR1 on the
   differences in E2 responses associated with uterine maturity. Our
   results yielded three major findings. First, we identified four
   intriguing DEGs that exhibited consistent changes. Second, we
   elucidated the role of EGR1 in the early, intermediate, and late E2
   responses. Finally, a similar E2 response was observed between
   EGR1-deficient mature OVX uteri and immature uteri in the late E2
   response.

   First, we investigated the expression patterns of four notable DEGs.
   These genes showed consistent downregulation up to 24 h (Arg2, Ripor2,
   and c-Kit) and upregulation up to 6 h (Bgn) after E2 treatment
   (Fig. [107]1C), suggesting that they are potential key targets of
   E2-mediated EGR1. ARG2, an arginase, catalyzes arginine metabolism and
   has been reported to regulate autophagy and senescence of immune and
   endothelial cells, as well as to facilitate successful fetal
   implantation during early pregnancy [[108]38–[109]41]. RIPOR2 is
   involved in T-cell migration and proliferation [[110]42]. c-KIT, which
   is induced by EGR1 in the uterus, plays a key role in determining cell
   fate by proliferation, differentiation, and apoptosis and is important
   for uterine receptivity [[111]36]. BGN, an extracellular proteoglycan,
   contributes to collagen fibrillogenesis in the uterine stroma
   [[112]43]. Its expression is regulated by E2 [[113]44] and important
   for connective tissue remodeling in non-pregnant premenopausal uteri
   [[114]45, [115]46]. Collectively, these DEGs, which are involved in
   immunity and reproduction, may provide insights into the implantation
   failure and impaired decidualization observed in Egr1 KO uteri.

   Second, we assessed the function of EGR1 in the early and intermediate
   E2 responses through cluster analysis of PPI networks based on DEGs and
   observed an upregulated IGF cluster in Egr1 KO uteri (Fig. [116]2C).
   IGF1-IGF1R signaling mediates E2-induced uterine proliferation [[117]7,
   [118]8, [119]47]. This IGF cluster includes Igf1r and Igf2. Increased
   expression of Igf1r can enhance IGF signaling activity. IGF2 acts as a
   major ligand for IGF1R and promotes proliferation similarly to IGF1
   [[120]48]. These results suggest that EGR1 regulates E2-ERα-IGF
   signaling activity, affecting uterine epithelial cell proliferation.

   Of the other clusters, the reproductive cluster was downregulated in
   Egr1 KO uteri (Fig. [121]2D). This cluster consists of Adamts-1, c-Kit,
   Inhbb, Inhba, Ror2, and Pdgfb. Adamts-1 and c-Kit are well-established
   targets of EGR1 and essential for endometrial receptivity and
   successful implantation [[122]35, [123]36]. Inhbb, Inhba, Ror2, and
   Pdgfb are essential for decidualization, endometrial repair following
   menstruation, crypt formation, implantation, and angiogenesis
   [[124]49–[125]52]. Given that Egr1 KO uteri exhibited implantation
   failure and compromised decidualization, these genes likely represent
   important targets for further investigations into the reproductive
   functions of EGR1.

   Pathway enrichment analysis of the late E2 response revealed that DNA
   metabolism and proliferation were upregulated in Egr1 KO uteri
   (Fig. [126]3A). Specifically, Parp1 and E2f1 were upregulated
   (Fig. [127]3B and Supplementary Table 1). PARP1 enhances the
   proliferative activity of E2F1 as a coactivator [[128]53–[129]55] and
   catalyzes PARylation, regulating DNA metabolism and mitosis [[130]28,
   [131]37]. Consistent with these roles, GSVA demonstrated increased
   activation of PARylation targets (Fig. [132]3C). These results suggest
   that hyperproliferation in Egr1 KO uteri is driven by both enhanced IGF
   clusters in the early and intermediate E2 responses and increased
   PARylation in the late E2 response.

   Conversely, immune-related pathways were downregulated in Egr1 KO uteri
   in the late E2 response (Fig. [133]3A). These pathways include H2-K1
   and H2-T23, which are associated with antigen expression; B2m, Ceacam1,
   Nlrp6, Arg1, and C3, which are important for the immune system; Lyz1
   and Lyz2, which contribute to antimicrobial activity; Ccl21a, Ccl21b,
   Ccl11, and Cxcl14, which are associated with chemokines. In contrast,
   “killing of cells of another organism” was consistently upregulated
   during the early and intermediate E2 responses (3 and 6 h;
   Fig. [134]2A). This pathway includes the E2-target gene Ltf and the
   immune cell-attracting chemokines Ccl21a, Ccl21b, Cxcl9, and Ccl8.
   Although immune score analysis did not reach statistical significance
   at 6 h (p = 0.07) or 24 h (p = 0.05; Fig. [135]3E), these results
   suggest that dysregulation of the immune system, which is tightly
   regulated by E2 for maintaining uterine receptivity [[136]56–[137]60],
   contributes to the infertility observed in Egr1 KO uteri.

   Finally, we found that the late E2 response in Egr1 KO uteri was
   similar to that in IM uteri, particularly in proliferation- and
   immune-related pathways (Fig. [138]4A and B). Although both mature OVX
   and IM uterus models have been widely used to study E2 responses, they
   exhibit distinct molecular mechanisms [[139]22]. In this context, the
   observed similarity at the pathway level between the two models is
   particularly intriguing. Gene-level analysis revealed only 16
   significantly altered genes common to both Egr1 KO and IM uteri
   compared to WT uteri (Fig. [140]4D). Among these genes, two are related
   to proliferation: Cdc25a, a cell cycle regulator, and Prim2, a DNA
   primase, and three are related to immune response: Fgl2, a mediator of
   inflammation [[141]61], Lgals3bp, and Trafd1. These results indicate
   that the similarity between Egr1 KO and IM uteri does not arise from
   identical underlying mechanisms; rather, distinct regulatory processes
   converge at the pathway level to yield a comparable E2 response.

   The limited gene-level similarity in the context of pathway-level
   similarity suggested two possible interpretations. One interpretation
   is that the anti-proliferative function of EGR1 is gradually
   established during uterine maturation. Although EGR1 is expressed in
   the immature uterus, its suppressive function remains to be fully
   established. Transcription factors (e.g., AP-1 and SP1) and epigenetic
   modifiers (e.g., histone acetyltransferases and DNA methyltransferases)
   may facilitate this progressive maturation of EGR1 function
   [[142]62–[143]65]. The second interpretation proposes that, despite
   normal EGR1 expression in immature uteri, the intrinsically high E2
   responsiveness may override EGR1’s anti-proliferative activity,
   resulting in hyperproliferation. In contrast, mature WT uteri, which
   express EGR1 and exhibit relatively lower E2 responsiveness, display
   properly regulated proliferation. However, in mature KO uteri lacking
   EGR1, this regulatory control is absent, leading to hyperproliferation
   even under lower E2 sensitivity. These interpretations are not mutually
   exclusive, and the observed transcriptomic similarity likely reflects
   both the incomplete establishment of EGR1’s anti-proliferative function
   and the inherently high E2 responsiveness of the immature uterus.
   Further studies are warranted to clarify their relative contributions,
   enhancing our understanding of uterine physiology and reproductive
   biology.

   While this study provides valuable insights into the function of EGR1
   as an E2 mediator in the uterus at the molecular level, there are
   limitations to consider. The number of samples for Egr1 KO and WT uteri
   at each time point was limited to only two, except for WT uteri at 3 h,
   which might have affected the statistical reliability of the Egr1 KO
   mice dataset. Although our findings delineate the primary mechanisms by
   which EGR1 mediates the E2 response, such as IGF signaling, PARylation,
   immune response, and reproductive processes, these conclusions are
   primarily based on transcriptomic analyses. Therefore, further in vitro
   and in vivo experimental data are needed to validate the biological
   significance of these findings. Addressing these gaps will enhance our
   understanding of the multifaceted roles of EGR1 in modulating the E2
   response and may in turn, drive significant advances in the
   understanding and treatment of infertility disorders.

Conclusions

   This study highlights the essential role of EGR1 in E2-mediated uterine
   receptivity. We demonstrate that E2-induced EGR1 regulates IGF and
   reproductive clusters in the early E2 response and governs DNA
   metabolism, proliferation, and the immune system in the late E2
   response. Notably, the late E2 response in mature OVX uteri lacking
   EGR1 closely resembles that of immature uteri. This similarity may
   result from either incomplete establishment of EGR1’s
   anti-proliferative function or the inherently high E2 responsiveness of
   the immature uterus. These results suggest that EGR1 is a pivotal
   mediator of E2-induced uterine epithelial proliferation via IGF
   signaling and plays a crucial role in regulating the reproductive
   cluster and immune response, which is essential for stable reproductive
   processes.

Supplementary Information

   [144]12864_2025_11904_MOESM1_ESM.xlsx^ (297.5KB, xlsx)

   Additional file 1: Supplementary Table 1. Differentially expressed
   genes (DEGs) between Egr1 knockout (KO) and wild-type (WT) uteri at
   each time point (3 h, 6 h, and 24 h). “Log2 Fold change” represents the
   logarithm (base 2) of the ratio of gene expression in Egr1 KO uteri
   compared to WT uteri. The “p-value” was calculated using limma. The
   “adj.p-value” represents the p-value adjusted using the
   Benjamini-Hochberg method.
   [145]12864_2025_11904_MOESM2_ESM.docx^ (144KB, docx)

   Additional file 2: Supplementary Figure 1. Protein-protein interaction
   networks based on differentially expressed genes at 3 and 6 h in Egr1
   knockout and wild-type uteri.
   [146]12864_2025_11904_MOESM3_ESM.xlsx^ (17.9KB, xlsx)

   Additional file 3: Supplementary Table 2. Pathway enrichment test for
   each cluster in Supplementary Figure 1A. “Category” indicates the data
   source for biological pathways, such as GO and KEGG. The “P-value” was
   calculated using an enrichment test. The “adj.p-value” is the adjusted
   p-value corrected for the multiple comparison problem. “Genes” are the
   genes overlapping with each enriched pathway.
   [147]12864_2025_11904_MOESM4_ESM.xlsx^ (18.5KB, xlsx)

   Additional file 4: Supplementary Table 3. Pathway enrichment test for
   each cluster in Supplementary Figure 1B. “Category” indicates the data
   source for biological pathways, such as GO and KEGG. The “P-value” was
   calculated using an enrichment test. The “adj.p-value” is the adjusted
   p-value corrected for the multiple comparison problem. “Genes” are
   genes overlapping with each enriched pathway.
   [148]12864_2025_11904_MOESM5_ESM.docx^ (125.3KB, docx)

   Additional file 5: Supplementary Figure 2. Hierarchical clustering
   results for a combined dataset of Egr1 knockout, wild-type, and
   immature mice samples.
   [149]12864_2025_11904_MOESM6_ESM.docx^ (292KB, docx)

   Additional file 6: Supplementary Figure 3. Heatmap of gene set
   signature scores computed by GSVA.
   [150]12864_2025_11904_MOESM7_ESM.docx^ (34.7KB, docx)

   Additional file 7: Supplementary Figure 4. The mRNA expression of Egr1
   in immature mouse uteri following treatment with E2 or AO.
   [151]12864_2025_11904_MOESM8_ESM.docx^ (358.1KB, docx)

   Additional file 8: Supplementary Figure 5. Unsupervised clustering
   results for genes associated with proliferation and immune response
   among Egr1 KO, WT, and IM samples for Fig. [152]4C.

Acknowledgements