Exosomal lncRNA OTUD6B-AS1 as a pathogenic nanocarrier promotes inflammatory macrophage polarization in endometritis via a targetable ceRNA circuit

Abstract

Endometritis, a major inflammatory cause of infertility, is driven by unresolved immune dysregulation in which macrophage polarization is critical. Yet, how inflammatory signaling is spatially propagated within the endometrial microenvironment remains unclear. Here, we report that exosomes—natural nanoscale extracellular vesicles—released from lipopolysaccharide (LPS)-injured endometrial epithelial cells (EECs) act as pathogenic nanocarriers that fuel macrophage-dependent inflammation. We demonstrate that LPS enhances exosome biogenesis and secretion via the AKT/ATG16L1 pathway. These exosomes efficiently deliver their molecular cargo to macrophages, triggering NF-κB activation and polarizing them toward a pro-inflammatory M1 phenotype. RNA sequencing identified lncRNA OTUD6B-AS1 as a highly enriched cargo in exosomes from inflamed EECs. Functional studies established that exosome-mediated transfer of lncRNA OTUD6B-AS1 is both necessary and sufficient to drive M1 polarization. Mechanistically, lncRNA OTUD6B-AS1 functions as a competing endogenous RNA (ceRNA), sequestering miR-128 to relieve its repression on Notch2, thereby amplifying NF-κB signaling. This axis was validated in clinical endometritis tissues, which exhibited elevated lncRNA OTUD6B-AS1 and Notch2 alongside reduced miR-128. Importantly, targeting this pathway—through genetic knockdown of lncRNA OTUD6B-AS1 or pharmacological inhibition of the miR-128/Notch2 node—abolished the pro-inflammatory effects. Our work not only delineates a new exosome-coordinated signaling circuit in endometritis but also highlights exosomes as druggable natural nanoparticles. These findings position exosome-based engineering—such as cargo modulation or designed vesicle delivery—as a promising nanomedicine strategy to intercept pathological cell-cell communication and treat inflammatory diseases.

Graphical abstract

1. Introduction

Endometritis poses a significant threat to fertility in mammals such as humans and dairy cows, primarily through inflammatory damage to the reproductive tract [1,2]. Statistically, approximately 20% of dairy cows suffer from persistent clinical endometritis at 3 weeks postpartum, while around 30% exhibit chronic subclinical endometritis [3]. Endometritis detrimentally impacts reproductive performance in multiple ways: it delays the resumption of postpartum ovarian cycles or prolongs the luteal phase, reduces conception rates, extends calving intervals, and decreases milk yield [[4], [5], [6]]. Endometritis is typically associated with postpartum uterine infections by pathogenic microorganisms, with common pathogens including Escherichia coli (E.coli) and Trueperella pyogenes [7]. As the first line of immune defense in the endometrium, endometrial epithelial cells detect the presence of pathogens and their virulence factors, such as lipopolysaccharide (LPS), via innate immune receptors (e.g., TLRs). This detection triggers the release of pro-inflammatory mediators, including IL-1β, IL-6 and TNF-α. These mediators subsequently recruit inflammatory cells such as macrophages to the site of infection and eliminate invading pathogens [8]. Consequently, a moderate inflammatory response facilitates pathogen clearance; however, an excessive and persistent response can lead to macrophage overactivation, initiating a cytokine storm that causes tissue damage [9].

Macrophages are a major source of pro-inflammatory cytokines within endometrial tissue and play a critical role during the endometrial inflammatory response [10]. Based on their activation state and functional differences, macrophages are broadly classified into two main types: classically activated M1 and alternatively activated M2 macrophages. These subtypes regulate the immune response, maintaining a dynamic balance between pro-inflammatory and anti-inflammatory reactions [11]. M1 macrophages typically exhibit Th1-like functions, promoting inflammation, apoptosis, and extracellular matrix (ECM) degradation through the release of pro-inflammatory cytokines and proteolytic enzymes, thereby contributing to chronic inflammation and tissue damage []. Conversely, M2 macrophages display Th2-like functions, resolving inflammation and promoting tissue remodeling and repair by secreting anti-inflammatory cytokines and ECM components [12]. During inflammatory responses in the endometrium, macrophages predominantly exhibit M1 polarization [13]. The balance between M1 and M2 macrophages is considered a key determinant in the progression towards endometrial inflammation or damage. However, how the inflammatory signals released by endometrial epithelial cells in response to LPS spread to macrophages and then induce endometrial inflammation is largely unclear.

Exosomes are extracellular vesicles with a diameter of 30-150 nm, featuring a double-layered lipid membrane structure and a density of 1.13-1.19 g/mL [14]. Almost all mammalian cells, including lymphocytes, neurons, tumor cells, and epithelial cells, can release exosomes. Exosomes mediate intercellular communication by transferring functional molecules such as proteins, DNA, and RNA, playing roles in numerous biological processes [15,16]. Exosomes also exert significant regulatory functions in pathological conditions such as endometrial infection and inflammation. Exosomes isolated from uterine lavage fluid of cows with endometritis significantly reduced blastocyst formation rates when co-incubated with embryos [17]. Critically, exosomes derived from endometrial epithelial cells have been demonstrated to modulate the function of endometrial immune cells; for instance, they promote the recruitment and activation of endometrial T lymphocytes, thereby mediating the occurrence and progression of endometritis. However, the regulatory effects of endometrial epithelial cells-derived exosomes on macrophage function remain unreported.

Long non-coding RNAs (lncRNAs) are transcripts longer than 200 nucleotides [18]. Once considered transcriptional “noise' without biological function, subsequent research has revealed that lncRNAs play important roles in various biological processes, including epigenetic regulation, cell cycle control, and immune signaling [19,20]. Through complete base pairing, lncRNAs can stabilize or promote mRNA translation, while partial base pairing can facilitate mRNA degradation or translational inhibition. Additionally, lncRNAs can function as competing endogenous RNAs (ceRNAs), competitively binding to microRNA (miRNAs) and consequently increasing the expression of miRNA target genes. MiRNA is a type of small noncoding RNA molecule that controls gene expression by directly binding to the 3′- UTR region of the target genes [21,22]. Exercise-induced miR-210 promotes cardiomyocyte proliferation and survival by inhibiting cyclin dependent kinase 10 expression, and mediates exercise-induced cardiac protection against ischemia/reperfusion injury [23]. LncRNAs and miRNAs play crucial roles in immune responses triggered by bacteria. For example, lncRNA XIST inhibited inflammation and apoptosis in mammary epithelial cells induced by E. coli and S. aureus via the NF‐κB/NLRP3 pathway [24]. lncRNA Gas5 promoted M1 polarization of macrophages in pneumonia patients through the miR-455-5p/SOCS3 pathway [25]. Nevertheless, studies on the role of lncRNAs in endometritis are scarce.

In the present study, we aim to elucidate the role and underlying mechanisms of endometrial epithelial cells-derived exosomes and their cargo lncRNAs in macrophages activation during endometritis. Our findings may reveal novel mechanisms and potential intervention strategies for endometritis.

2. Materials and methods

2.1. Animal sample collection

The study protocol was approved by the Life Scientific Ethic Committee of Yunnan Agricultural University (approval no. 202203053). Endometrial tissues were collected from healthy and endometritis-affected postpartum Holstein cows (n = 5; parity: median = 5, range = 4–6; postpartum period: 21–60 days) selected from local dairy farms. The health status of the cows was diagnosed by experienced veterinarians based on previously established criteria: (1) presence of purulent vaginal discharge and (2) visual inflammatory lesions in the endometrial tissue. Histopathological staining was performed as previously described [26]. Briefly, the collected tissues were fixed overnight in 4% paraformaldehyde, embedded in paraffin, sectioned and stained. Histopathological changes in the endometrium were observed under an optical microscope (Olympus, Japan).

2.2. Endometrial epithelial cell and macrophage culture

Uterine tissue samples from healthy cows were transported on ice and processed within 1 h. Primary endometrial epithelial cells (EECs) were isolated and cultured following a previously described protocol [27]. Briefly, uterine specimens were washed three times with phosphate-buffered saline (PBS), and segments (<2 cm) near the oviduct were dissected, opened longitudinally, and rinsed with Dulbecco's PBS. The exposed endometrial tissue was digested overnight with 1% pronase (Sigma, USA). The digested tissue was gently scraped, and the resulting cell suspension was washed three times with PBS and centrifuged (300×g, 4 °C, 5 min). EECs isolated were seeded in DMEM medium (HyClone, USA) supplemented with 15% (HYcezmbio, Wuhan, China) and 100 IU/mL penicillin-streptomycin, then incubated at 37 °C under 5% CO₂. EECs were identified using a specific antibody against cytokeratin 18. The macrophage cell line (BoMac) is a transformed macrophage cell line originally described by Stabel and Stabel [28], and the cells were maintained in DMEM supplemented with 10% fetal bovine serum.

2.3. Isolation and characterization of EECs-Derived exosomes

Exosomes were isolated from the culture supernatants of control and LPS-injured EECs following a previously established protocol [29]. Briefly, supernatants were sequentially centrifuged at 2000×g for 20 min and 10,000×g for 30 min to remove death cells, cell debris and large vesicles. The resultant supernatant was then filtered through a 0.22 mm filter (Millipore, USA) to eliminate any remaining larger particles.

Exosomes were subsequently pelleted by ultracentrifugation at 100,000×g for 70 min at 4 °C. The crude exosome pellets were resuspended in sterile PBS and subjected to a second round of ultracentrifugation at 100,000×g for 70 min at 4 °C to further reduce contaminating proteins and lipoproteins. The final purified exosomes from LPS-treated EECs (LPS-Exo) and control EECs (Control-Exo) were used for subsequent experiments. To exclude the possibility of LPS carryover, the endotoxin levels in the final exosome preparations were measured using a chromogenic limulus amebocyte lysate (LAL) assay.

For transmission electron microscopy (TEM), exosomes were fixed in 2% paraformaldehyde, applied to carbon-coated grids, and negatively stained with 3% phosphotungstic acid. Morphology and size were examined using a Hitachi H-600 microscope (Japan). Exosome concentrations and particle size distribution were determined by Nanoparticle Tracking Analysis (NTA) using a Nanosight NS300 particle size analyzer. For flow cytometry, exosomes were stained with a FITC-conjugated anti-CD81 antibody (BD Pharmingen™, USA) as an exosomal marker and analyzed on a BD Biosciences flow cytometer (USA). Exosomal markers (CD81 and TSG101) were further confirmed by western blotting.

2.4. Exosome labeling

Exosomes were labeled using a exosome labeling and tracing kit (Beyotime, China) following the manufacturer's protocols. Briefly, exosome pellets were incubated with PKH67 dye (100 μL) for 5 min. Labeling was terminated by adding an equal volume of fluorescent labeling termination solution, followed by ultracentrifugation and PBS wash to remove excess free dye and non-exosomal dye aggregates. Labeled exosomes were then immediately used for uptake assays.

2.5. Exosomal RNA isolation and lncRNA sequencing

Exosomal RNA from control EECs (n = 3) and LPS-stimulated EECs (n = 3) was extracted using an exoRNeasy Kit (QIAGEN, Germany). RNA integrity, purity, and concentration were assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA), NanoPhotometer® spectrophotometer (IMPLEN, USA), and Qubit® RNA Assay Kit (Life Technologies, USA), respectively. High-quality RNA was sent to RiboBio (Guangzhou, China) for library preparation and Illumina-based sequencing.

Putative lncRNA–miRNA and miRNA-mRNA interactions were predicted using miRanda, PITA, and RNAhybrid. Competing endogenous RNA (ceRNA) networks were constructed based on shared miRNA binding sites. KEGG and GO enrichment analyses of target genes were performed with a significance threshold of p < 0.05.

2.6. Cell treatments

• (1)To establish an in vitro model of EEC injury, EECs were treated with 100 μg/mL LPS (E. coli O111:B4; Sigma, USA) for 24 h. Apoptosis was confirmed by PI staining.
• (2)To assess exosome uptake, macrophages were incubated with PKH-67-labeled exosomes in exosome-free medium for 24 h and visualized under a fluorescence microscope. To evaluate effects of exosome on macrophage activation, macrophages were co-cultured with Control-exo or LPS-exo for 24 h. For all functional experiments involving exosome treatment, exosome doses were normalized to equal particle numbers. This normalization ensured that observed differences in macrophage responses reflected qualitative differences in exosome cargo, rather than quantitative differences in exosome yield.
• (3)To determine the role of ATG16L1 in exosome release, EECs were transfected with ATG16L1-targeting siRNA (si-ATG16L1, 50 nM, 12 h) before LPS treatment.
• (4)To examine the role of lncRNA OTUD6B-AS1 in macrophage activation, macrophages were transfected with lncRNA OTUD6B-AS1 overexpression plasmids or siRNA, followed by LPS stimulation.
• (5)To test whether lncRNA OTUD6B-AS1 acts via miR-128, macrophages were co-transfected with lncRNA OTUD6B-AS1 overexpression plasmids and miR-128 mimic before LPS treatment.
• (6)To assess Notch2 involvement, macrophages were pretreated with the Notch2 inhibitor DAPT for 1 h before transfection with lncRNA OTUD6B-AS1 overexpression plasmids or miR-128 inhibitor and LPS stimulation.

2.7. Co-culture experiments

Co‐culture of EECs and macrophages was achieved by Transwell inserts (Corning, USA), with EECs in the upper layer and macrophages in the lower layer. The 0.4 μm porous membrane between the two layers allowed exosomes to pass freely. Prior to co-culture, EECs were treated with LPS for 24 h, the medium was removed, and the cells were then washed twice with PBS, and supplemented with fresh medium, followed by co‐culture with macrophages in the lower layer for an additional 24 h. GW4869 (Sigma, USA) was added to the EEC culture at a concentration of 10 μM to restrain the release of exosomes from EECs.

2.8. Propidium iodide (PI) staining

After treatment, EECs were fixed with 4% paraformaldehyde, and apoptosis was assessed using a PI staining kit according to the manufacturer's instructions.

2.9. RNA fluorescence in situ hybridization (FISH)

The Cy5 labeled lncRNA specific probe was purchased from GenePharma and the signal was detected using the RNA FISH kit (GenePharma, China) according to the manufacturer's instructions.

2.10. miRNA mimic, inhibitor, siRNA, and transfection

miR-128 mimic, miR-128 inhibitor, lncRNA OTUD6B-AS1 siRNA, and overexpression plasmids were synthesized by GenePharma (Shanghai, China). Transfection was performed via electroporation following the manufacturer's protocols.

2.11. Plasmid construction and luciferase reporter assay

Wild-type (WT) lncRNA OTUD6B-AS1 or Notch2 3′-UTR sequences containing miR-128 binding sites were cloned into the psiCHECK™-2 vector (Promega, USA). Mutant (MUT) plasmids were generated by site-directed mutagenesis. HEK293T cells were co-transfected with reporter plasmids and miR-128 mimic or controls. Luciferase activity was measured after 24 h using the Dual-Luciferase Reporter Assay System (Promega, USA).

2.12. RNA isolation and quantitative PCR (qPCR)

Total RNA was extracted with TRIzol (Vazyme, China), reverse-transcribed, and analyzed by qPCR using PerfectStart Green qPCR SuperMix (TRANS, China) on a real-time PCR system (Analytikjena, Germany). GAPDH and U6 served as internal controls. Relative expression was calculated using the 2^^(−ΔΔCt) method. Primer sequences are provided in Table S1.

2.13. RNA pull-down and RNA immunoprecipitation (RIP) assays

The relationship between lncRNA OTUD6B-AS1 and miR-128 was further verified by RNA pull-down assay. The binding RNA was purified by TRIzol, and qPCR was applied for the determination of lncRNA OTUD6B-AS1 enrichment. RIP kit (Millipore, USA) was applied for detecting the binding of lncRNA OTUD6B-AS1 and Ago2. The antibody applied in RIP assay was rabbit anti-Ago2, and rabbit anti-IgG was used as a control.

2.14. Immunofluorescence assay

Cells or tissue sections were permeabilized with 0.1% Triton X-100, blocked with 10% BSA, and incubated with primary antibodies at 4 °C overnight. After incubation with fluorescent secondary antibodies for 1 h at room temperature, nuclei were stained with DAPI, and images were captured using a fluorescence microscope (SOPTOP, China). The antibodies used include iNOS (Proteintech, Cat^#18985-1-AP), Arg1 (Proteintech, Cat^#18985-1-AP), Goat Anti-Rabbit IgG (Bioss, Cat^#bs-0295G) and Goat Anti-Mouse IgG (Bioss, Cat^#bs-0296Gs).

2.15. Western blotting

Total protein was extracted from tissues or cells using a commercial kit (Vazyme, China). Western blotting was performed as previously described [30]. In brief, proteins were separated by SDS-PAGE, transferred to PVDF membranes, and probed with antibodies against appropriate primary antibodies. HRP-conjugated secondary antibodies were applied, and signals were detected using an Amersham ImageQuant 800 system (Cytiva, USA). The antibodies used include TSG101 (Genuin Biotech, Cat^#51942), CD81 (Santa Cruz Biotechnology, Cat^#sc-18877), AKT (abcam, Cat^#AB81283), Phospho-AKT (abcam, Cat^#AB179463), NF-κB p65 (abcam, Cat^#AB32536), Phospho-NF-κB p65 (Cell Signaling Technology, Cat^#3033S), Notch2 (Santa Cruz Biotechnology, Cat^#sc-51869), RBP-Jκ (Santa Cruz Biotechnology, Cat^#sc-271128), and β-actin (Bioss, Cat^#bs-0061R).

2.16. Statistical analysis

All quantitative data are expressed as the mean ± standard error of the mean (SEM). Normality of data distribution was assessed, and homogeneity of variances was verified. For comparisons between two independent groups, an unpaired, two-tailed Student's t-test was applied. Comparisons among three or more groups were performed using one-way analysis of variance (ANOVA), followed by multiple comparisons. Statistical analyses were conducted using GraphPad Prism version 10.0 (GraphPad Software, San Diego, CA, USA). A p-value less than 0.05 was considered statistically significant, with the following notation used in figures: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

3. Results

3.1. Analysis of macrophage activation in uterine tissues of cows with endometritis

Macrophage activation, particularly pro-inflammatory M1 polarization, is recognized as a key factor in the pathogenesis of various inflammatory diseases. To further confirm whether macrophages are activated during the process of endometrial inflammatory injury, we examined the expression levels of iNOS (an M1 marker) and Arg1 (an M2 marker) in the uterine tissues of affected cows. Immunofluorescence staining revealed a significant increase in iNOS (Fig. 1A and B) and a decrease in Arg1 (Fig. 1C and D) in the endometritis group, indicating a predominant M1 polarization of macrophages. Moreover, qPCR results showed an elevated mRNA level of iNOS (Fig. 1E) and reduced expression of Arg1 (Fig. 1F), accompanied by upregulation of the pro-inflammatory cytokines IL-1β, IL-6 and TFN-α (Fig. 1G–J), further validating the immunofluorescence findings. These results suggest that pro-inflammatory macrophage activation and enhanced inflammatory responses occur in the uterine tissues of endometritis.

Fig. 1.

Analysis of macrophage activation in the uterine tissue of cows with endometritis. (A, B) Representative immunofluorescence (IF) staining images (A) and quantitative analysis (B) of iNOS (M1 marker, red) in endometrial tissues from healthy cows and cows with endometritis. Nuclei were counterstained with DAPI (blue). (C, D) Representative IF staining images (C) and quantitative analysis (D) of Arg1 (M2 marker, red) in endometrial tissues. (E, F) Relative mRNA expression levels of iNOS (E) and Arg1 (F) in endometrial tissues, as determined by qPCR. (G, H) Relative expression levels of IL-1β, IL-6 and TNF-α in endometrial tissues, as determined by IHC. (I, J) Relative mRNA expression levels of IL-1β (I) and IL-6 (J) in endometrial tissues, as determined by qPCR. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.2. Isolation and identification of endometrial epithelial cells (EECs)-Derived exosomes

Exosomes are important mediators of intercellular communication in inflammation. Studies have shown that exosomes derived from EECs can regulate the function of endometrial T lymphocytes; however, the role of exosomes from inflamed EECs in macrophage activation associated with endometritis remains unclear. In this study, primary EECs were isolated via enzymatic digestion and mechanical separation. Cell morphology and integrity were confirmed by immunofluorescence staining for cytokeratin 18 (Fig. 2A). Subsequently, LPS derived from E. coli was used to induce inflammatory injury in EECs. PI staining indicated that LPS stimulation significantly induced an increase in the number of apoptotic cells (Fig. 2B and C) and upregulated IL-1β and IL-6 levels (Fig. 2D and E) compared to the control group. Exosomes were then extracted from the culture supernatant of injured EECs via ultracentrifugation (Fig. 2F). TEM images showed exosomes with a classic cup-shaped morphology and a diameter ranging from 30 to 150 nm (Fig. 2G). NTA further confirmed that the size distribution of the isolated particles was consistent with that of exosomes (Fig. 2H). Notably, the particle concentration of exosomes derived from LPS-stimulated EECs (LPS-exo) was significantly higher than that of exosomes from control EECs (Control-exo) (Fig. 2I), indicating enhanced exosome secretion under inflammatory conditions. Western blotting and flow cytometry analysis further showed enrichment of the exosomal markers TSG101 and CD81 in the exosome fraction (Fig. 2J and K). In addition, the results of LAL assay indicated that endotoxin contamination was barely detectable in the exosome preparations (Fig. 2L). These data confirm the successful isolation of exosomes from LPS-injured EECs.

Fig. 2.

Isolation and characterization of exosomes derived from endometrial epithelial cells (EECs). (A) Immunofluorescence staining for cytokeratin 18 (CK18, red) confirming the epithelial identity of isolated primary EECs. Nuclei were stained with DAPI (blue). (B, C) Representative images (B) and quantitative analysis (C) of PI staining in EECs treated with or without LPS. (D, E) Relative expression levels of IL-1β (D) and IL-6 (E) in control and LPS-stimulated EECs. (F) Schematic diagram of exosome isolation from EEC culture supernatant via ultracentrifugation. (G) Representative transmission electron microscopy (TEM) image of isolated exosomes. (H, I) Nanoparticle tracking analysis (NTA) showing the size distribution (H) and particle concentration (I) of exosomes derived from control EECs (Control-exo) and LPS-stimulated EECs (LPS-exo). (J) Western blotting analysis of the exosomal markers TSG101 and CD81 in isolated exosomes. (K) Flow cytometry analysis confirming the presence of CD81 on isolated exosomes. (L) Analysis of the endotoxin levels in isolated exosomes using a chromogenic limulus amebocyte lysate (LAL) assay. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.3. LPS enhances exosome release from EECs via the AKT/ATG16L1 pathway

To explore the molecular mechanism underlying increased exosome release upon LPS stimulation, we examined the expression of several key genes related to exosome biogenesis and secretion in EECs, including SNAP23, Rab27a, Rab7a, Rab5, ATG5, and ATG16L1 (Fig. 3A). The results showed that LPS stimulation significantly upregulated ATG5, Rab5 and ATG16L1, with ATG16L1 exhibiting the most pronounced increase (Fig. 3B). Functional screening through individual gene knockdown in EECs revealed that while depletion of each gene impaired exosome secretion, ATG16L1 knockdown caused the most significant reduction in exosome concentration (Fig. 3C), underscoring its prominent and critical role in mediating LPS-induced exosome release. Recent studies have shown that TNF-α upregulates ATG16L1 via AKT activation to promote exosome release from mesenchymal stem cells [31]. Therefore, we further examined p-AKT level and found that LPS stimulation significantly increased p-AKT content in EECs (Fig. 3D). The activation of the PI3K/AKT pathway has been confirmed to lead to phosphorylation of downstream transcription factor cAMP response element binding protein (CREB) [32,33]. According to the database of AnimalTFDB transcription factor binding site prediction, it has been determined that CREB exhibits high binding affinity to the promoter region of ATG16L1 (Fig. 3E), implying that AKT may regulate the transcription of ATG16LI through CREB. To directly test whether AKT activation is required for the enhanced exosome secretion observed under inflammatory conditions, we treated LPS-stimulated EECs with the AKT inhibitor LY294002. AKT inhibition not only significantly downregulated ATG16L1 expression (Fig. 3F), but also reduced the particle concentration of exosomes released from LPS-stimulated EECs (Fig. 3G). Intriguingly, overexpression of ATG16L1 partially rescued the reduction in exosome release caused by AKT inhibition (Fig. 3G). Collectively, these results indicate that LPS may enhance exosome release via the AKT/ATG16L1 pathway.

Fig. 3.

LPS enhances exosome release from EECs via the AKT/ATG16L1 pathway. (A) Schematic diagram of exosome biogenesis and secretion. (B) Relative expression levels of genes related to exosome biogenesis and secretion in control and LPS-stimulated EECs. (C) The concentration of exosomes released from ATG5, ATG16L1 or Rab5-knockdown EECs, as measured by NTA. (D) Western blotting analysis of phosphorylated AKT (p-AKT) and total AKT protein levels in control and LPS-stimulated EECs. (E) Binding analysis of transcription factor CREB and ATG16L1 promoter. (F) Relative mRNA expression level of ATG16L1 in EECs treated with LPS and the AKT inhibitor LY294002. (G) The concentration of exosomes released by EECs after transfection with ATG16L1 overexpression plasmids (OE-ATG16L1). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

3.4. Exosomes from LPS-stimulated EECs induce pro-inflammatory macrophage activation

To investigate exosome transfer between EECs and macrophages, we labeled EECs-derived exosomes with the fluorescent dye PKH67 and co-cultured them with macrophages (Fig. 4A). Fluorescently labeled exosomes were taken up by macrophages in vitro (Fig. 4B). Notably, compared to exosomes from control EECs, those from LPS-stimulated EECs significantly increased the phosphorylation of the pro-inflammatory transcription factor NF-κB p65 (p-p65) in macrophages (Fig. 4C). This was accompanied by increased iNOS (Fig. 4D–E) and decreased Arg1 expression (Fig. 4F–G). Furthermore, we performed additional co-culture experiments using a Transwell system. Specifically, LPS-injured or control EECs were cultured in the upper chamber, while macrophages were cultured in the lower chamber, allowing for the exchange of soluble factors including exosomes—without direct cell–cell contact (Fig. 4H). To specifically assess the role of exosomes in this intercellular communication, we treated co-cultures with or without GW4869, a well-established inhibitor of exosome release. As shown in Fig. 4I and J, co-culture with LPS-injured EECs obviously increased iNOS and decreased Arg1 expression in macrophages, and this effect was substantially attenuated by GW4869 treatment. These results suggest that exosomes from LPS-injured EECs are a novel pro-inflammatory mediator that can induce M1 polarization in macrophages.

Fig. 4.

Exosomes from LPS-stimulated EECs induce pro-inflammatory macrophage activation. (A) Schematic diagram of the experimental setup for exosome uptake. (B) Fluorescence microscopy images showing the uptake of PKH67-labeled exosomes (green) by macrophages. Cytoskeleton was stained with Phalloidin (red), and nuclei were stained with DAPI (blue). (C) Western blotting analysis of phosphorylated NF-κB p65 (p-p65) in macrophages treated with Control-exo or LPS-exo. (D, E) Representative immunofluorescence (IF) staining images (D) and quantitative analysis (E) of iNOS (greed) in macrophages. (F, G) Representative IF staining images (F) and quantitative analysis (G) of Arg1 (red) in macrophages. (H) Schematic diagram of co-culture experiments. (I, J) Relative mRNA expression levels of iNOS (I) and Arg1 (J) in macrophages after co-culture with EECs. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.5. Transcriptomic profiling reveals significant enrichment of lncRNA OTUD6B-AS1 in exosomes derived from LPS-stimulated EECs

To determine which active molecules in exosomes are responsible for inducing macrophage activation, we performed the exosomal RNA degradation assay. LPS-exo were treated with RNase A alone or in combination with the membrane-permeabilizing detergent Triton X-100 for 4 h, and then co-incubated with macrophages. As shown in Fig. 5A and B, Macrophages exposed to exosomes pretreated with both RNase A and Triton X-100 exhibited markedly lower iNOS expression and higher Arg1 expression compared to those treated with RNase A alone, demonstrating that the pro-inflammatory capacity of LPS-exo depends on both vesicle integrity and the functional delivery of internal RNA cargo. LncRNAs play important roles in macrophage activation by targeting miRNAs. To identify specific exosomal lncRNAs mediating macrophage activation, we compared the lncRNA expression profiles of Control-exo and LPS-exo using next-generation sequencing (Fig. 5C). A total of 232 lncRNAs were differentially expressed in LPS-exo, with 118 upregulated (p < 0.05, log2 ^{(fold change)} > 1) (Fig. 5D). GO enrichment analysis of these lncRNAs highlighted significant associations with endometritis-related biological processes, including regulation of inflammatory response and cell proliferation (Fig. 5E). KEGG pathway analysis further identified Notch signaling pathway related to immune regulation and cell activation (Fig. 5F). Among the differentially expressed exosomal lncRNAs, we selected six upregulated lncRNAs for qPCR validation. The expression abundance and log2 ^{(fold change)} of six differentially sorted lncRNAs are shown in Table S2. As depicted in Fig. 5G, lncRNA OTUD6B-AS1 (LOC100848906) was significantly upregulated in LPS-exo (approximately 3-fold higher than in Control-exo). Meanwhile, an additional mini-screen experiment showed that only knockdown of lncRNA OTUD6B-AS1 significantly attenuated LPS-exo-induced iNOS expression, whereas knockdown of LOC112446643 or LOC107132829 had no significant effect (Fig. S1). More importantly, treatment with RNase A alone failed to degrade exosomal lncRNA OTUD6B-AS1, whereas combined RNase A and Triton X-100 treatment resulted in its marked degradation (Fig. 5H), confirming that the functional lncRNA OTUD6B-AS1 is encapsulated within intact exosomes. Altogether, these results provide direct evidence that exosomes secreted by inflamed EECs promote pro-inflammatory macrophage activation in an exosomal lncRNA-dependent manner. Additionally, we constructed a ceRNA network based on the differentially expressed lncRNAs. This network suggested that the lncRNA OTUD6B-AS1/miR-128/Notch2 signaling axis plays a potential immunomodulatory role (Fig. 5I). LncRNA OTUD6B-AS1 was significantly upregulated in LPS-stimulated EECs (Fig. 5J) and was primarily localized in the cytoplasm (Fig. 5K and L). Consistent with the expression pattern observed following LPS stimulation, EECs stimulated with inactivated E. coli also exhibited significantly elevated lncRNA OTUD6B-AS1 expression compared to unstimulated controls (Fig. S2). Further analysis showed that lncRNA OTUD6B-AS1 expression was significantly higher (Fig. 5M–O), while miR-128 expression was lower (Fig. 5P), in the endometrial tissues of cows with endometritis compared to healthy controls. Concurrently, the expression levels Notch2, RBP-Jκ and p-p65 were significantly upregulated in the endometritis group (Fig. 5Q).

Fig. 5.

Transcriptomic profiling reveals significant enrichment of lncRNA OTUD6B-AS1 in exosomes derived from LPS-stimulated EECs. (A, B) LPS-exo was treated with RNase A alone or in combination with Triton X-100 for 4 h, and then co-incubated with macrophages. Relative expression levels of iNOS (A) and Arg1 (B) in macrophages. (C) Schematic overview of the RNA sequencing and analysis workflow. (D) Volcano plot showing differentially expressed lncRNAs in LPS-exo compared to Control-exo. (E, F) Gene Ontology (GO) biological process enrichment analysis (E) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis (F) of the differentially expressed lncRNAs. (G) qPCR validation of the 6 upregulated lncRNAs in Control-exo and LPS-exo. (H) LPS-exo was treated with RNase A alone or in combination with Triton X-100 for 4 h, and then co-incubated with macrophages. Relative mRNA expression level of lncRNA OTUD6B-AS1 in macrophages. (I) A proposed competing endogenous RNA (ceRNA) network involving lncRNA OTUD6B-AS1, miR-128, and Notch2. (J) Relative mRNA expression level of lncRNA OTUD6B-AS1 in control and LPS-stimulated EECs. (K, L) RNA fluorescence in situ hybridization (RNA-FISH) showing the subcellular localization of lncRNA OTUD6B-AS1 (red) in EECs (K) and its quantitative cytoplasmic/nuclear distribution (L). Nuclei were stained with DAPI (blue). (M–P) Relative mRNA expression levels of lncRNA OTUD6B-AS1 (M − O) and miR-128 (P) in endometrial tissues from healthy cows and cows with endometritis, as determined by qPCR (O, P) and RNA-FISH (M) with quantification (N). (Q) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 protein levels in endometrial tissues. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.6. EECs-derived exosomes induce pro-inflammatory macrophage activation via delivery of lncRNA OTUD6B-AS1

To determine whether lncRNA OTUD6B-AS1 can be transferred from EECs to macrophages, we measured its expression in macrophages after co-culture with EECs-derived exosomes. Macrophages treated with LPS-exo exhibited higher lncRNA OTUD6B-AS1 level than those treated with Control-exo (Fig. 6A), which was further confirmed by RNA-FISH (Fig. 6B and C). To investigate the role of lncRNA OTUD6B-AS1 in macrophage activation, we overexpressed or knocked it down in macrophages. The transfection efficiency was then confirmed by qPCR, and the results showed that transfection of lncRNA OTUD6B-AS1 overexpression plasmids (OE-lncRNA) significantly increased lncRNA OTUD6B-AS1 expression in macrophages compared to control plasmids (OE-NC) (Fig. 6 D), confirming successful overexpression. Overexpression of lncRNA OTUD6B-AS1 promoted the expression of Notch2, RBP-Jκ, p-p65 (Fig. 6E) and iNOS (Fig. 6F and G), while reducing Arg1 expression (Fig. 6H and I). Conversely, knockdown of lncRNA OTUD6B-AS1 reduced the expression of Notch2, RBP-Jκ, p-p65 (Fig. 6J) and iNOS (Fig. 6K and L) expression, and increased Arg1 expression (Fig. 6M and N).

Fig. 6.

EECs-derived exosomes induce pro-inflammatory macrophage activation via delivery of lncRNA OTUD6B-AS1. (A) Relative mRNA expression level of lncRNA OTUD6B-AS1 in macrophages treated with Control-exo or LPS-exo. (B, C) RNA-FISH images (B) and quantitative analysis (C) showing lncRNA OTUD6B-AS1 (red) transfer to macrophages after co-culture with Control-exo or LPS-exo. Nuclei were stained with DAPI (blue). (D) Relative mRNA expression level of lncRNA OTUD6B-AS1 in macrophages after transfection with lncRNA OTUD6B-AS1 overexpression plasmids (OE-lncRNA) or control plasmids (OE-NC). (E–I) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 (E), along with immunofluorescence (IF) quantitative analysis of iNOS (F, G) and Arg1 (H, I) protein levels in macrophages after transfection with OE-lncRNA or OE-NC. (J–N) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 (J), along with IF quantitative analysis of iNOS (K, L) and Arg1 (M, N) protein levels in macrophages after lncRNA OTUD6B-AS1 knockdown (si-lncRNA) or control treatment (si-NC). (O) Relative mRNA expression level of lncRNA OTUD6B-AS1 in exosomes isolated from lncRNA OTUD6B-AS1-knockdown LPS-stimulated EECs (si-lncRNA-LPS-exo) or exosomes from siRNA NC-transfected LPS-stimulated EECs (si-NC-LPS-exo). (P–S) IF quantitative analysis of iNOS (P, Q) and Arg1 (R, S) protein levels in macrophages treated with si-lncRNA-LPS-exo or si-NC-LPS-exo. (T) Relative mRNA expression levels of iNOS and Arg1 in macrophages treated with exosomes isolated from control EECs overexpressing lncRNA OTUD6B-AS1 (OE-lncRNA-Control-exo) or exosomes from control plasmids-transfected EECs (OE-NC-Control-exo). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

To further verify whether lncRNA OTUD6B-AS1 mediates the effect of EECs-derived exosomes on macrophage activation, we knocked down lncRNA OTUD6B-AS1 in LPS-stimulated EECs, isolated exosomes from these cells, and co-cultured them with macrophages. As shown in Fig. 6O, exosomes isolated from lncRNA OTUD6B-AS1-knockdown LPS-stimulated EECs (si-lncRNA-LPS-exo) exhibited significantly reduced lncRNA OTUD6B-AS1 expression compared to those from siRNA NC-transfected LPS-stimulated EECs (si-NC-LPS-exo), confirming successful engineering of exosomes with specific lncRNA cargo depletion. Of particular interest, no significant differences were observed in particle concentration between these two groups of exosomes (Fig. S3). Functionally, compared to the si-NC-LPS-exo group, exosomes from lncRNA OTUD6B-AS1-knockdown LPS-stimulated EECs significantly decreased iNOS (Fig. 6P and Q) and increased Arg1 (Fig. 6R and S) expression, indicating attenuated M1 polarization. Conversely, exosomes isolated from control EECs overexpressing lncRNA OTUD6B-AS1 (OE-lncRNA-Control-exo) markedly promoted M1 polarization upon co-incubation with macrophages, as evidenced by increased iNOS and decreased Arg1 expression, compared to exosomes from control plasmids-transfected EECs (OE-NC-Control-exo) (Fig. 6T). Collectively, these results demonstrate that exosomes derived from LPS-injured EECs induce pro-inflammatory macrophage activation specifically through the transfer of lncRNA OTUD6B-AS1.

3.7. lncRNA OTUD6B-AS1 acts as a ceRNA by sponging miR-128 to facilitate pro-inflammatory macrophage activation

Macrophages treated with LPS-exo exhibited lower miR-128 expression than those treated with Control-exo (Fig. 7A). Luciferase reporter assays showed that miR-128 mimic significantly reduced luciferase activity from the WT-lncRNA OTUD6B-AS1 reporter plasmids but not the MUT plasmids (Fig. 7B). RNA pull-down assay verified that lncRNA OTUD6B-AS1 was significantly enriched in the miR-128 pull-down complex compared to the biotin-labeled NC control (Fig. 7C), confirming the direct binding between lncRNA OTUD6B-AS1 and miR-128. Furthermore, to investigate whether lncRNA OTUD6B-AS1 and Notch2 coexist in the RNA-induced silencing complex (RISC), we performed Ago2-RIP assays in macrophages, and the results showed that the lncRNA OTUD6B-AS1's specific sponge of Ago2 ascended markedly compared to IgG control (Fig. 7D). Additionally, overexpression of lncRNA OTUD6B-AS1 in macrophages significantly decreased miR-128 expression (Fig. 7E). To investigate whether miR-128 mediates the regulatory effect of lncRNA OTUD6B-AS1 on macrophage activation, we co-transfected macrophages with lncRNA OTUD6B-AS1 overexpression plasmids and miR-128 mimic. The results showed that overexpression of miR-128 significantly reversed the lncRNA OTUD6B-AS1-induced increase in Notch2, RBP-Jκ, p-p65 (Fig. 7F) and iNOS (Fig. 7G and H) expression, and alleviated the decrease in Arg1 expression (Fig. 7I and J). In addition, we also analyzed the expression of pro-inflammatory cytokines IL-1β and IL-6. The results demonstrated that overexpression of lncRNA OTUD6B-AS1 significantly upregulated IL-1β and IL-6 mRNA levels in macrophages compared to the OE-NC group, whereas miR-128 overexpression significantly attenuated this upregulation (Fig. 7K and L). These results indicate that lncRNA OTUD6B-AS1 promotes pro-inflammatory macrophage activation by sponging miR-128.

Fig. 7.

lncRNA OTUD6B-AS1 acts as a ceRNA by sponging miR-128 to facilitate pro-inflammatory macrophage activation. (A) Relative mRNA expression level of miR-128 in macrophages treated with Control-exo or LPS-exo. (B) Luciferase reporter assay in HEK293T cells co-transfected with wild-type (WT) or mutant (MUT) lncRNA OTUD6B-AS1 reporter plasmids and miR-128 mimic or mimic NC. (C) RNA pull-down detection of the enrichment of miR-128 to lncRNA OTUD6B-AS1. (D) Ago2 RIP assay analysis of the enrichment of lncRNA OTUD6B-AS1 pulled-down from the Ago2 protein. (E) Relative mRNA expression level of miR-128 in macrophages transfected with OE-NC or OE-lncRNA. (F–J) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 (F), along with immunofluorescence (IF) quantitative analysis of iNOS (G, H) and Arg1 (I, J) protein levels in macrophages co-transfected with OE-lncRNA and miR-128 mimic or mimic NC. (K, L) Relative mRNA expression levels of IL-1β (K) and IL-6 (L) in macrophages co-transfected with OE-lncRNA and miR-128 mimic or mimic NC. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

3.8. Notch2 mediates the regulatory effect of the lncRNA OTUD6b-AS1/miR-128 axis on macrophage activation

Several databases predicted that Notch2 is the potential target of miR-128 (Fig. 8A), implying that miR-128 exerts a regulatory effect on macrophage activation through Notch2. Meanwhile, macrophages treated with LPS-exo exhibited higher Notch2 expression than those treated with Control-exo (Fig. 8B). Moreover, overexpression of lncRNA OTUD6B-AS1 significantly increased Notch2 protein level (Fig. 8C), suggesting a potential regulatory role for Notch2 in exosomal lncRNA OTUD6B-AS1-induced macrophage activation. To determine whether Notch2 mediates this effect, we inhibited Notch2 using the inhibitor DAPT in macrophages overexpressing lncRNA OTUD6B-AS1. Notch2 inhibition significantly blocked the lncRNA OTUD6B-AS1-induced increase in Notch2, RBP-Jκ, p-p65 (Fig. 8D) and iNOS (Fig. 8E and F) expression, and alleviated the decrease in Arg1 expression (Fig. 8G and H). These results indicate that lncRNA OTUD6B-AS1 promotes pro-inflammatory macrophage activation by enhancing Notch2 expression.

Fig. 8.

Notch2 mediates the regulatory effect of the lncRNA OTUD6B-AS1/miR-128 axis on macrophage activation. (A) Predictive analysis of miR-128 targets using multiple databases. (B) Western blotting analysis of Notch2 protein levels in macrophages treated with Control-exo or LPS-exo. (C) Western blotting analysis of Notch2 protein levels in macrophages transfected with OE-NC or OE-lncRNA. (D–H) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 (D), along with immunofluorescence (IF) quantitative analysis of iNOS (E, F) and Arg1 (G, H) protein levels in macrophages treated with OE-NC or OE-lncRNA and the Notch2 inhibitor DAPT. (I) Luciferase reporter assay in HEK293T cells co-transfected with WT or MUT Notch2 3′UTR reporter plasmids and miR-128 mimic or mimic NC. (J, K) Relative protein (J) and mRNA (K) expression levels of Notch2 in macrophages transfected with miR-128 mimic or mimic NC. (L–P) Western blotting analysis of Notch2, RBP-Jκ, and p-p65 (L), along with IF quantitative analysis of iNOS (M, N) and Arg1 (O, P) protein levels in macrophages co-treated with miR-128 inhibitor or inhibitor NC and DAPT. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Furthermore, the luciferase activity was decreased in HEK293T cells co-transfected with WT-Notch2 3′-UTR luciferase reporter plasmids and miR-128 mimic (Fig. 8I), suggesting that the 3′-UTR of Notch2 is directly targeted by miR-128. Overexpression of miR-128 significantly reduced Notch2 mRNA and protein levels (Fig. 8J and K), further indicating that Notch2 is a key target of miR-128. Additionally, inhibition of miR-128 with an inhibitor promoted Notch2, RBP-Jκ, p-p65 (Fig. 8L), increased iNOS (Fig. 8M and N), and decreased Arg1 (Fig. 8O and P). However, the inhibition of Notch2 by DAPT alleviated these effects of miR-128. These results indicate that miR-128 regulates pro-inflammatory macrophage activation by targeting Notch2.

4. Discussion

Disruption of endometrial epithelial barrier integrity and infiltration of inflammatory cells such as pro-inflammatory macrophages are hallmark pathological features of endometritis. The molecular mechanisms through which the initial damage to endometrial epithelial cells (EECs) drives inflammatory cascades and exacerbates endometrial tissue damage remain incompletely understood. This study identified a novel signaling axis: exosome-carried lncRNA OTUD6B-AS1 mediates intercellular communication between EECs and macrophages by regulating the miR-128/Notch2 pathway, thereby aggravating inflammatory damage in endometrial tissue (Fig. 9). This finding not only deepens our understanding of the role of exosomes in the pathogenesis of endometritis but also provides a molecular basis for developing new therapeutic strategies.

Fig. 9.

A proposed model illustrating the exosome-mediated lncRNA OTUD6B-AS1/miR-128/Notch2 axis in aggravating endometritis. Upon LPS-induced damage, endometrial epithelial cells (EECs) release increased exosomes carrying elevated levels of lncRNA OTUD6B-AS1. These exosomes are taken up by endometrial macrophages. The transferred lncRNA OTUD6B-AS1 acts as a molecular sponge to sequester miR-128, leading to the derepression and upregulation of its target gene, Notch2. The enhanced Notch2 signaling subsequently promotes macrophage polarization towards a pro-inflammatory M1 phenotype, characterized by increased NF-κB activation and iNOS expression, thereby exacerbating endometrial inflammation and tissue damage.

Intrauterine infection by Gram-negative bacteria is a major cause of endometritis, with pathogenic E. coli playing a prominent role [7,34]. LPS, a key component of the Gram-negative bacterial cell wall, activates the TLR4/NF-κB signaling pathway and initiates the transcription of pro-inflammatory cytokines, contributing critically to E. coli-induced endometritis [27]. Consistent with previous reports, we observed significantly elevated levels of pro-inflammatory factors and a marked increase in pro-inflammatory macrophages in the endometrial tissue of cows with endometritis. Similar phenomena have been documented in humans [35], mice [36], and mares [37]. As central effector cells in innate immunity, macrophages eliminate pathogens through phagocytosis and cytokine secretion [38]. However, excessive activation can lead to the overproduction of inflammatory mediators and exacerbate tissue damage. EECs constitute the first line of defense against pathogens. Under uncontrolled inflammatory conditions, they secrete elevated levels of inflammatory factors and chemokines, further amplifying self-damage and promoting inflammatory progression. Nevertheless, the relationship between injured EECs and pro-inflammatory macrophage activation requires further investigation.

Exosomes, nanoscale vesicles secreted by various cell types including epithelial cells, serve as novel mediators of intercellular communication and play roles in regulating inflammation and immune responses in inflammatory diseases [39,40]. For instance, renal tubular epithelial cell-derived exosomes promote inflammation in tubulointerstitial pathology [41], prompting us to explore whether exosomes from damaged EECs regulate macrophage polarization. Due to the difficulty of directly isolating EECs-derived exosomes from the uterine tissue of diseased animal, we established an in vitro model of EEC injury using LPS derived from pathogenic E. coli (O111:B4). High concentrations of LPS induced strong inflammatory responses and apoptosis in EECs, successfully mimicking the damaged state. Notably, we observed that LPS-stimulated EECs released a greater number of exosomes. Exosome biogenesis is a multistage process encompassing endocytosis, early endosome formation, progression to late endosomes/multivesicular bodies (MVBs), and ultimately MVB fusion with the plasma membrane for exosome release [42]. To investigate the mechanism by which LPS stimulation increases exosome secretion from EECs, we initially examined the expression of six well-documented regulators of exosome biogenesis and secretion—SNAP23, Rab27a, Rab7a, Rab5, ATG5, and ATG16L1 [[43], [44], [45]]. Our screening revealed that LPS stimulation upregulated Rab5, ATG5 and ATG16L1 to varying degrees, with ATG16L1 exhibiting the most significant increase. Furthermore, functional knockdown experiments revealed that ATG16L1 depletion led to the most pronounced reduction in exosome concentration, suggesting that while multiple components contribute to LPS-induced exosome release, ATG16L1 may play a particularly important role in this process. Importantly, ATG16L1 is a well-known component of the autophagy machinery, and recent studies have unveiled significant molecular interplay between autophagy and exosome biogenesis. Critically, emerging evidence demonstrates that autophagy-related proteins can regulate exosome production through non-canonical, autophagy-independent mechanisms. For instance, Guo et al. demonstrated that the ATG5-ATG16L1 complex promotes exosome production by de-acidifying multivesicular bodies via dissociation of the V1V0-ATPase, independent of canonical macroautophagy [44]. This finding supports the concept that ATG16L1 can directly influence exosome biogenesis through a mechanisms separable from autophagic flux. The PI3K/AKT pathway is known to modulate autophagy in various cell types [46]. In our study, we observed that LPS activates AKT, and AKT inhibition suppressed ATG16L1 expression. Given the established links between AKT signaling, autophagy regulation, and the emerging roles of ATG16L1 in exosome biogenesis, the alterations in exosome secretion under LPS stimulation may indeed reflect broader changes in cellular autophagic flux. However, the precise mechanistic relationship between autophagy status and enhanced exosome release in LPS-stimulated EECs requires further investigation.

Previous studies have firmly established that activation of the PI3K/AKT pathway leads to phosphorylation of various downstream transcription factors, including CREB [32,33]. CREB is a well-known transcriptional regulator that binds to specific DNA sequences to promote gene expression [47]. Based on this well-documented signaling cascade, we hypothesized that AKT might regulate ATG16L1 transcription through CREB. To explore this possibility, we performed a bioinformatic analysis using the AnimalTFDB transcription factor binding profile database, defined transcription factor binding sites. Our analysis identified putative CREB binding sites within the promoter region of the ATG16L1 gene, suggesting that ATG16L1 is a potential transcriptional target of CREB. This prediction provides a plausible molecular link whereby LPS-activated AKT could induce CREB phosphorylation and nuclear translocation, leading to CREB binding to the ATG16L1 promoter and subsequent transcriptional upregulation of ATG16L1, thereby enhancing the release of EECs-derived exosomes. Notably, the conclusion was also strongly supported by a recently published study demonstrating that in TNF-α-preconditioned infrapatellar fat pad mesenchymal stem cells, activation of the PI3K/AKT pathway leads to upregulation of ATG16L1, which subsequently facilitates exosome secretion [31]. Collectively, these findings establish that AKT activation is necessary for the LPS-triggered secretion of exosomes derived from EECs, and positions AKT upstream of ATG16L1 in this regulatory cascade. Nevertheless, further investigation is warranted to elucidate the precise molecular interplay between AKT and ATG16L1.

Macrophages have been found to efficiently internalize exogenous exosomes and alter their functional status [48,49]. In the present study, exosomes from LPS-stimulated EECs were taken up by macrophages, and acted as a novel pro-inflammatory mediator to promote macrophages polarization towards the M1 pro-inflammatory phenotype. Notably, this exosome-mediated regulation of macrophage polarization is not unique to endometritis; similar findings have been reported in the context of rotator cuff tendon-bone healing, where exosomes were also shown to modulate macrophage phenotype and influence tissue regeneration [50]. Additional co-culture experiments performed using a Transwell system, combined with the exosome release inhibitor GW4869, further demonstrated that exosomes released from inflamed EECs play a critical and non-redundant role in driving pro-inflammatory macrophage activation. This finding aligns with recent studies showing that exosomes from LPS-treated EECs impair trophoblast development [51]. A recent research report suggests that extracellular vesicles containing miR-29a-3p accumulate in tumors and downregulate collagen composition by targeting COL1A1, implying that extracellular vesicles can regulate receptor cell function by delivering bioactive molecules [21]. Importantly, numerous studies have highlighted the regulatory roles of exosomal lncRNAs in various inflammatory diseases. For example, cholangiocyte-derived exosomal lncRNA H19 promotes macrophage activation and subsequent liver inflammation under cholestatic conditions [52]. However, the role of exosomal lncRNAs in endometritis has not been reported. To identify key lncRNAs mediating exosome-induced pro-inflammatory macrophage activation, we performed lncRNA expression profiling of exosomes from LPS-treated EECs. The results revealed multiple differentially expressed lncRNAs in exosomes from LPS-stimulated EECs, which were significantly enriched in immune regulation and cell activation pathways such as the Notch signaling pathway. Further validation showed that lncRNA OTUD6B-AS1 was significantly upregulated in both LPS-stimulated EECs and their exosomes, suggesting its involvement in endometritis pathogenesis. Moreover, knockdown of lncRNA OTUD6B-AS1 weakened LPS-induced apoptosis and secretion of pro-inflammatory cytokines in EECs (Fig. S4), suggesting that this lncRNA indeed plays a pro-inflammatory role within the donor cells themselves, contributing to the initial injury response of EECs upon LPS challenge. Similarly, upregulation of lncRNA OTUD6B-AS1 expression was also observed in EECs stimulated by inactivated E. coli. These results demonstrate that our LPS-induced inflammatory model recapitulates key molecular changes triggered by clinically relevant pathogens, supporting the validity and clinical relevance of our findings. Encouragingly, macrophages co-cultured with exosomes from LPS-stimulated EECs also exhibited increased lncRNA OTUD6B-AS1 levels. Functional experiments further demonstrated that overexpression of lncRNA OTUD6B-AS1 in macrophages enhanced NF-κB p65 activation and M1 marker iNOS expression, further indicating its pro-inflammatory role. Based on these findings, we hypothesized that EEC-derived exosomes induce macrophage activation via transferring lncRNA OTUD6B-AS1. To test this, we knocked down lncRNA OTUD6B-AS1 in EECs, isolated the exosomes, and then co-cultured them with macrophages. As expected, lncRNA OTUD6B-AS1 knockdown reversed the pro-inflammatory effects of exosomes from LPS-stimulated EECs on macrophages. These results imply that EEC-derived exosomes induce pro-inflammatory activation of macrophages by delivering the key lncRNA OTUD6B-AS1 into macrophages. The exosomes secreted from the LPS‐treated EECs contained cytokines, RNAs, and other active molecules, and they could also directly induce M1 polarization in macrophages. After degradation of the exosomal RNA by RNase A and Triton X‐100 synchronously, the ability of LPS-exo to induce macrophage activation was greatly counteracted. Altogether, these results provide direct evidence that exosomes secreted by inflamed EECs promote pro-inflammatory macrophage activation in an exosomal lncRNA-dependent manner.

LncRNAs function as competing endogenous RNAs (ceRNAs) by binding to miRNAs and play critical roles in inflammatory responses [53]. However, there are few reports on the role and mechanism of lncRNA in endometritis. Through ceRNA network analysis, we identified the lncRNA OTUD6B-AS1/miR-128/Notch2 network as a potential key regulator of pro-inflammatory macrophage activation. More importantly, endometrial tissue from cows with endometritis showed significantly increased lncRNA OTUD6B-AS1 expression, decreased miR-128 levels, and upregulated p-p65 and Notch2 expression, further confirming the potential role of the lncRNA OTUD6B-AS1/miR-128/Notch2 network in endometritis. Similarly, macrophages co-cultured with exosomes from LPS-stimulated EECs exhibited reduced miR-128 and elevated Notch2 levels. Further analysis revealed binding sites between lncRNA OTUD6B-AS1 and miR-128, and overexpression of miR-128 abolished the lncRNA OTUD6B-AS1-induced promotion of p-p65 and iNOS expression in macrophages, indicating that lncRNA OTUD6B-AS1 regulates macrophage activation by sponging miR-128. Previous studies reported that miR-128 negatively regulates S. aureus-induced inflammatory responses in murine macrophages [54], supporting its role in lncRNA OTUD6B-AS1-mediated macrophage activation. Subsequent investigations confirmed Notch2 as a direct target of miR-128. Another study in amniotic epithelial cells has shown that miR-128 can target JAG1 to inhibit the activation of the Notch pathway [55]. Recent studies have shown that the Notch signaling pathway is a key regulator of macrophage function [56]; RBP-Jκ-dependent Notch 2 activates NF-κB pathway, promoting M1 macrophage polarization and exacerbating inflammation [57]. To determine whether Notch2 mediates the regulatory effects of lncRNA OTUD6B-AS1 and miR-128 on macrophage activation, we inhibited Notch2 using the specific inhibitor DAPT. Notch2 inhibition significantly attenuated the pro-inflammatory effects induced by lncRNA OTUD6B-AS1 overexpression and miR-128 knockdown.

Admittedly, while our findings clearly establish exosomal lncRNA OTUD6B-AS1 as a key mediator of inflammatory cross-talk, we recognize that the endometrial microenvironment is complex and likely involves additional inflammatory factors that warrant further investigation.

5. Conclusion

In conclusion, this study reveals a novel mechanism by which exosomes from inflamed EECs deliver lncRNA OTUD6B-AS1 to macrophages, regulating the miR-128/Notch2 pathway and promoting pro-inflammatory macrophage activation, thereby driving endometrial tissue damage. The identification of the exosome/lncRNA OTUD6B-AS1/miR-128/Notch2 signaling axis provides new theoretical insights and potential therapeutic targets for endometritis.

CRediT authorship contribution statement

Jing Yang: Investigation, Validation, Writing – original draft. Yajing Chen: Data curation, Investigation, Validation. Kui Wang: Investigation, Methodology, Validation. Lifang Mi: Investigation, Validation. Yushan Chen: Investigation, Validation. Xiaobing Li: Investigation, Validation. Gefen Yin: Funding acquisition, Supervision, Writing – review & editing. Kangfeng Jiang: Conceptualization, Funding acquisition, Supervision, Writing – review & editing.

Declaration of competing interest

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

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.