Animals

Mature Kunming White outbred strain mice (aged 6–8 weeks) were procured from the Experimental Animal Center of Chongqing Medical University and maintained in a regulated setting with a 14-h light phase and a 10-h dark phase. The experimental protocols involving animals were ethically approved by the Committee of Ethics for Animal Experiments at Chongqing Medical University (No. IACUC-CQMU-2017–0227). To induce pregnancy, Female mice were mated with male mice of the identical genetic background, and the detection of a vaginal plug was considered indicative of day 1 (D1) of gestation. On D5 of pregnancy, following the intravenous injection administration of 125 μL of 0.4% trypan blue (Sigma-Aldrich, St. Louis, MO), the implantation sites exhibited a blue coloration, distinguishing them from the inter-implantation sites. Subsequently, five mice were euthanized for the collection of endometrial samples.

Patients and tissue specimens

Total of 6 preeclampsia patients and 6 normal pregnant women were enrolled in this study. Pregnant women with preeclampsia were randomly recruited from the Chongqing Health Center for Women and Children in 2021. The inclusion criteria for the two groups were as follows: Normal pregnancy: Singleton pregnant women exhibiting normal blood pressure levels. Preeclampsia: Patients with systolic blood pressure ≥ 160 mmHg and diastolic ≥ 110 mmHg on at least two occasions, 6 h apart, after 20 weeks of gestation. Diagnosis was made during the gestational period ranging from 27 to 36 weeks. Exclusion criteria encompassed multiparous pregnancies, pre-existing hypertension, kidney disorders, gestational diabetes mellitus, fetal chromosomal or congenital abnormalities, and instances of stillbirth. All study subjects recruited in the study provided written informed consent before the cesarean section. The study was approved by the Ethics Committee of Chongqing Health Center for Women and Children (NO. yjkt20211213). Table S2 displays the clinical profiles of preeclampsia patients and the control group.

HTR8/SVneo cell culture

The human extravillous trophoblast-like cell line HTR8/SVneo was maintained in RPMI-1640 medium enriched with 5% fetal bovine serum and 1% penicillin/streptomycin and then maintained at 37 °C in a humidified incubator with 5% CO2.

Isolation and in vitro decidualization of primary mouse uterine endometrial stromal cell (mESCs)

mESCs were derived and cultured using established protocols (Chen et al. 2023). Decidualization for mESCs was artificially induced in vitro by 10 nM estrogen (Sigma, USA) and 1 μM progesterone (Sigma, USA). mESCs were stimulated for 24–72 h and then collected for further experimental procedures.

In vivo decidualization

Mature Kunming White outbred strain mice was used to establish artificial decidualization model. The mouse artificial decidualization model was established as previously described (Long et al. 2019). Briefly, on the morning of 4th day of pseudopregnancy, artificial decidualization was initiated by intraluminally injecting 10 μL of corn oil into one side of the uterine horn (ID), while the contralateral uterine horn was lef untreated and used as the control (IDC). Uterine tissues were then harvested on the morning of day 8 of pseudopregnancy for subsequent analyses.

Tandem mass tag (TMT) based proteomic analysis

TMT-based proteomic analysis was conducted by Applied Protein Technology located in Shanghai, China. The proteins extracted from the treated cells were first separated on an SDS-PAGE gel and then underwent enzymolysis through filter-aided sample preparation (FASP digestion), as referenced in (Hughes et al. 2019). A total of 100 μg of peptide mixture per sample was labeled with TMT reagent strictly following the manufacturer's guidelines provided by Thermo Fisher Scientific, USA. To explore the various pathways modulated by circ-Hdac4, a pathway analysis was carried out utilizing the Kyoto Encyclopedia of Genes and Genomes database (https://www.genome.jp/kegg/pathway.html) and Gene Ontology (https://www.blast2go.com/).

Public RNA-seq data

Public RNA-seq data were obtained from the Gene Expression Omnibus (GEO) database (accession number: GSE2361). We performed expression profiling of 36 types of normal human tissues. Detailed sample information and Affymetrix.CEL files are available at http://www.genome.rcast.u-tokyo.ac.jp/normal/.

siRNA and mimic transfection

The circ-Hdac4 interfering RNA (circ-Hdac4 siRNA) was custom designed and synthesized by Shanghai GenePharma Co., Ltd. (Shanghai, China). Similarly, the miR-30c mimic was crafted and synthesized by Guangzhou RiboBio Co., Ltd. According to previous experimental procedures (Su et al. 2022), circ-Hdac4 siRNA and the miR-30c mimic were diluted to concentration of the usage solution is 60 nm (circ-Hdac4 siRNA) and 100 nm (miR-30c mimic). Both the circ-Hdac4 siRNA and the miR-30c mimic were then introduced into stromal cells using HiPerFect transfection reagent (QIAGEN, Germany). Following a 6-h transfection period, the stromal cells underwent in vitro decidualization with estrogen and progesterone treatment for 48 h. Afterward, the transfected cells were harvested for subsequent examination. The sequences for the circ-Hdac4 siRNA were as follows: siRNA-1: 5'-GCCATCCAGGTTCATCTCT-3', siRNA-2: 5'-CAAAGCCATCCAGGTTCAT-3', and siRNA-3: 5'-CCCAAAGCCATCCAGGTTC-3'.

Construction of adenovirus/lentivirus and cell transfections

The sequences of circ-Hdac4 siRNA and the RBPJ mRNA CDS region were submitted to Hanheng Biotechnology Co., Ltd. (located in Shanghai, China) for the custom design and synthesis of a circ-Hdac4 interfering adenovirus (Ad-circ-Hdac4-shRNA), an RBPJ overexpression lentivirus (LV-RBPJ-OE), as well as their corresponding negative controls: adenovirus (Ad-NC-shRNA) and lentiviral vector control (LV-vector). The MOI (multiplicity of infection) values for Ad-circ-Hdac4-shRNA and LV-RBPJ-OE were determined to be 50 pfu/cell and 30 pfu/cell, respectively. The appropriate volumes of viral fluid were dispensed into the 24 well plates. Following a 6-h incubation period, the stromal cells underwent in vitro decidualization treatment with estrogen and progesterone for durations of 24, 48, and 72 h, respectively. Afterward, the culture medium was collected and preserved at −20℃ for subsequent analysis. Fresh medium was then replenished in the well plates.

Transwell invasion assay

The invasion capabilities of HTR-8 cells cell invasiveness were evaluated using a transwell chamber assay. The medium, previously used to culture stromal cells, was collected. HTR-8 cells suspended in this recovered medium, containing either Ad-circ-Hdac4-shRNA or Ad-NC-shRNA, were seeded at a density of 5 × 104 cells/chamber into a 12.0 μm transwell chamber (NEST, WuXi, China) that had been pretreated with Matrigel (BD Matrigel™ Basement Membrane Matrix, USA). Following a 48-h incubation period, the cells remaining on the upper membrane of the transwell chamber were gently wiped away. The HTR-8 cells that had migrated through the membrane were then fixed using cold methanol for 10 min andthen dyed with 0.1% crystal violet for 30 min.

Real-time PCR

Following the protocol of manufacturer, total RNA was isolated by utilizing RNAiso Plus Reagent (TaKaRa, Dalian, China). The PrimeScriptTM RT Kit (TaKaRa) facilitated the reverse transcription of circRNA and mRNA. For the reverse transcription of miRNA, the Mir-XTM miRNA First-Strand Synthesis Kit was employed. Primer sequences utilized in this research are provided in Table S3. Relative gene expression levels were analyzed using the 2-ΔΔCt method. β-actin was chosen as the internal control for mRNA and circRNA quantifications, whereas U6 served as the reference for miRNA quantification.

Histology and immunohistochemistry

Uterine tissue was preserved using 4% paraformaldehyde (Biosharp, HeFei, China), dehydrated through a graded alcohol series and then embedded in paraffin (Solarbio, BeiJing, China). Five-micrometer sections were cut and prepared for further examination. The slides were dewaxed, rehydrated, and stained using standard hematoxylin and eosin staining (Nanjing Jiancheng, NanJing, China) techniques to assess general tissue morphology and identify various cell types. For immunohistochemical antigen retrieval, the sections were boiled in EDTA buffer at PH 8.0 (Zhongshan Biosciences, Beijing, China). Following this, the sections were treated with the primary antibodies and incubated overnight at 4℃ specified in Table S4. They were then exposed to biotinylated secondary antibodies (Zhongshan Biosciences, BeiJing, China) for 1 h at 37℃. Negative controls were set up by substituting IgG for the primary antibodies. Positive signals were visualized using 3,3′-diaminobenzidine, and cell nuclei were counterstained with hematoxylin (Nanjing Jiancheng, NanJing, China). Finally, the images were gathered and evaluated using a photomicroscope (Olympus, Japan).

Cell fluorescence analysis

mESCs were isolated and grown on slides. Following fixation in 4% paraformaldehyde for 15 min, the sections were incubated with a blocking solution of 5% BSA. Subsequently, they were treated with the primary antibodies detailed in Table S4. Afterward, the sections were exposed to fluorescein isothiocyanate (FITC)-labeled rabbit IgG (Beyotime Biotechnology, Nanjing, China) for 1 h at 37 °C in a darkened environment. Nuclei were subsequently counterstained with DAPI (Beyotime Biotechnology). In the final step, the sections were imaged and analyzed via confocal fluorescence microscopy (Nikon, Japan).

Western blot

Tissue of mouse uterus and cultured cell proteins were extracted according to the manufacturer's instructions (Beyotime Biotechnology, Nanjing, China) as follows. Mouse uterine tissues and cultured cells were homogenized in ice-cold lysis buffer combined with protease/phosphatase inhibitors (Beyotime Biotechnology, Nanjing, China). The homogenates were incubated on ice for 30 min to facilitate complete protein solubilization. Subsequently, the lysates were centrifuged for 15 min at 12,000 × g (12,000 rpm) and 4 °C. The supernatant was gathered and combined with 5X SDS-PAGE loading buffer (Beyotime Biotechnology, Nanjing, China) in a 4:1 volume ratio (supernatant: buffer). The mixture was then denatured by heating at 100 °C for 10 min. The protein samples were then subjected to SDS-PAGE gel electrophoresis to separate proteins based on molecular weight. Then target protein was transferred to PVDF at 250 mA. The membranes were blocked with 5% BSA (Solarbio, Beijing, China), then incubated with the primary antibodies listed in Table S4 overnight at 4℃. Afterward, the membranes were exposed to a secondary antibody (Cell Signaling Technology, CA, USA) conjugated to horseradish peroxidase. The immunoreactive signal was detected using ECL reagent (Millipore, Darmstadt, Germany), and the results were imaged using the ChemiDocTM XRS + system (Bio-Rad).

In situ hybridization

Frozen sections were fixed in 4% paraformaldehyde at 25℃ for 1 h, subsequently treated with 1% Triton for 20 min. Hybridization was then carried out at 42℃ for 16 h using a digoxin-labeled circ-Hdac4 probe, which was tailor-made and synthesized by Guangzhou Jisai Biotechnology Co., Ltd. (Guangzhou, China). The specific sequence of the circ-Hdac4 probe was 5'-GCAGAGATGAACCTGGATGGCTTT-3'. Following hybridization, the sections were treated with an anti-digoxygenin antibody bound to alkaline phosphatase (1:5000, Roche, Germany) and stained accordingly. A positive in situ hybridization signal appeared as dark brown. In the final step, the image was captured and analyzed using a photomicroscope (Olympus, Japan).

Dual-luciferase activity assay

The Dual-Luciferase Activity Assay was carried out in partnership with Hanheng Biotechnology Co., Ltd (Shanghai, China). The predicted binding sites for circ-Hdac4 and miR-30c, as well as miR-30c and RBPJ, together with a mutant sequence devoid of the binding site, were inserted into a pSI-Check2 plasmid vector (Promega, Beijing, China). For transfection, HEK293T cells were plated in 24-well plates at 70–80% confluency. The plasmid constructs (1 μg/well) and miR-30c mimic (50 nM final concentration) were co-transfected exploring Lipofectamine 3000 reagent (Thermo Fisher, Massachusetts, USA) in adherence to the manufacturer’s protocol. Briefly, plasmid DNA and miRNA mimic were diluted in Opti-MEM medium (Thermo Fisher, Massachusetts, USA), mixed with Lipofectamine 3000 at a 1:2 DNA:Lipofectamine ratio (w/v), and incubated at room temperature for 15 min to facilitate transfection complexes. The complexes were then added to cells and incubated for 6 h before replacing the medium with fresh DMEM containing 10% FBS. After a 48-h incubation period, the activities of Firefly and Renilla luciferase were monitored using the Promega Dual-Luciferase System. The HEK293T cells were lysed using passive lysis buffer (Promega, Beijing, China), and subsequently, Luciferase Assay Reagent II (Promega, Beijing, China) was added to the cell lysate for the quantification of Firefly luciferase activity, followed by the addition of Stop & Glo Reagent (Promega, Beijing, China).

Uterine horn injection

The method has previously been described in detail (Long et al. 2019). On pregnant day 4 (D4), pregnant mice were anesthetized, and a midline abdominal incision was made to reveal the uterine horns. The left uterine horn received intraluminal injection containing either 60 nM circ-Hdac4 siRNA or 100 nM miR-30c mimic complexed with HiPerFect Transfection Reagent (QIAGEN, Germany) in sterile PBS. The right horn served as an internal control, injected with scrambled siRNA (or negative control mimic) mixed equivalently with HiPerFect. Post-injection, needles were retained for 10 s to prevent reflux. Surgical incisions were closed with absorbable sutures. Mice were monitored until recovery and euthanized at pregnant day 8 (D8). Uterine tissues were harvested for further investigations.

Isolation of nuclear and cytoplasmic RNA

Separation of the nuclear and cytoplasmic fractions was carried out according to (Ding et al. 2019) using the RNA Subcellular Isolation Kit (Active Motif, 25,501). The procedure was as follows: after three washes with ice-cold PBS, the cells were spun at 300 × g for 5 min, and the resulting pellet was resuspended in the Complete Lysis Buffer, and incubated on ice for 10 min. The cell samples were then centrifuged at 500 × g for 5 min at 4 °C to obtain the nuclear and cytoplasmic fractions, from which subcellular RNA was subsequently extracted. The extracted RNA was used for RT‑qPCR analysis, and the primer sequences are presented in Table S3.

Reduced uterine perfusion pressure (RUPP)

Following uterine horn injection on the fifth day of pregnancy, RUPP (Reduced Uterine Perfusion Pressure) was induced as previously described (Colson et al. 2023). In summary, pregnant mice were anesthetized using isoflurane, and their breathing was assisted with a rodent ventilator (supplied by Harvard Apparatus, USA). A 2 cm abdominal incision was meticulously performed, and four silver clips were precisely positioned around the vascular arcades linked to the ovarian and uterine vessels. Blood pressure was measured in conscious mice using tail-cuff plethysmography (from Visitech System, USA) on conscious mice. On the 18th day of pregnancy, blood samples and tissues were collected for subsequent experimental procedures.

Statistical analysis

Statistical analyses were conducted utilizing the SPSS 22.0 software package. The T-test was used to assess significant differences between two groups, whereas Comparisons among multiple groups were analyzed using one-way ANOVA, with statistical significance set at p < 0.05. The data are expressed as means ± standard deviation.