Human placentae were obtained immediately from deliveries with written informed consent from participating women under a protocol [Protocol No. (2013) N025] approved by the Ethics Committee of Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University. Modes of deliveries included uncomplicated term pregnancies after elective cesarean section without labor (term non-labor, TNL); spontaneous labor at term (term labor, TL); preterm pregnancies terminated by cesarean section without labor for maternal or fetal conditions including placenta previa, vasa previa, and fetal distress (preterm non-labor, PNL); and spontaneous preterm labor with no indication of infection (preterm labor, PL). Pregnancies with complications such as preeclampsia, fetal growth restriction, and gestational diabetes were excluded. Within 20 min of delivery, the placentae were transported in cold normal saline to the laboratory and the reflected fetal membranes were cut from the placenta, and the amnion layer was peeled off from the chorion layer. For RNA and protein extraction, the amnion layer was cut within 5 cm of the spontaneous or artificial rupture site and rinsed with ice-cold normal saline to remove residual blood and then placed in liquid nitrogen for storage at −80°C until RNA and protein extraction. RNA extracted from the amnion in the TNL and TL groups was processed for transcriptomic sequencing for the initial investigation of the bradykinin system in amnion, which has been reported in a previous publication of our laboratory (28). RNA extracted from the amnion in the TNL, TL, PNL, and PL groups was also processed for measurements with quantitative real-time-polymerase chain reaction (qRT-PCR) to confirm the changes of the bradykinin system in labor observed with transcriptomic sequencing. Protein extracted from the amnion in the TNL, TL, PNL, and PL groups was used for the measurements of bradykinin changes in labor with enzyme-linked immunosorbent assay (ELISA). To investigate the distribution of BDKRB1 and BDKRB2 in human amnion, amnion tissue at the rupture site from TL patients (n = 3) was processed for staining with immunohistochemistry. To study the effects of BK, DABK, LPS, and SAA1 on the expression of PTGS2, BDKRB1, or BDKRB2 and the underlying mechanisms, fibroblasts were isolated from the entire amnion of the reflected membranes collected from the cesarean section at term without labor. The number of women from whom the fetal membranes were collected was given in the corresponding figure legend for each study. Demographic and clinical features of pregnant women are given in Tables 1 and 2 .

To compare the amounts of bradykinin peptides in amnion with or without labor at term and preterm, amnion tissue pieces were cut 5 cm within the rupture site of the amnion and then ground in liquid nitrogen. The ground tissue was homogenized and lysed in ice-cold RIPA lysis buffer containing a protease inhibitor cocktail and centrifuged. Protein in the supernatant was collected for bradykinin peptide measurement with an ELISA kit (Abcam, Cambridge, MA, USA, #ab136936) with no distinction between BK and DABK, following the protocol provided by the manufacturer.

Tissue sections were cut from paraffin-embedded human amnion tissue from TL after fixing in 4% paraformaldehyde. After quenching the endogenous peroxidase activity with 0.3% H₂O₂ and blocking with normal serum, primary antibodies against human BDKRB1 (Proteintech, Wuhan, China, #26672-1-AP) and BDKRB2 (R&D System, Minneapolis, MN, USA, #MAB9434-SP) were applied to the section at 1:200 dilutions for overnight incubation at 4°C. After washing with PBS, a secondary antibody conjugated with biotinylated horseradish peroxidase was applied to the section for further incubation for 1 h at room temperature. The red color was then developed using the substrate 3-amino-9-ethyl carbazole (Vector Laboratories, Burlingame, CA, USA). The slide was counterstained with hematoxylin (blue color) and examined under a regular light microscope (Zeiss, Oberkochen, Germany).

Amnion fibroblasts were isolated from the amnion collected from TNL as described previously (35). Briefly, the amnion tissue was digested twice with 0.125% trypsin (Sigma Chemical Co., St. Louis, MO, USA) and then washed thoroughly with normal saline to remove epithelial cells for culture. The remaining tissue was digested with 0.1% collagenase (Sigma) to isolate fibroblasts, and fibroblast cells in the digestion medium were collected by centrifugation. Isolated epithelial cells and fibroblasts were cultured at 37°C in 5% CO₂–95% air in DMEM containing 10% fetal calf serum (FCS) plus 1% antibiotics (all from Life Technologies Inc., Grand Island, NY, USA) for 3 days before treatments. The identity of cells was verified by staining for mesenchymal cell marker vimentin. More than 95% of the cells were positive for vimentin in amnion fibroblast preparation and the epithelial cells were pure and negative for vimentin.

Three days after plating, the culture medium was replaced with DMEM free of phenol red and FCS for treatments of amnion fibroblasts. To investigate the effects of BK and DABK on the expression of PTGS2, time course and concentration-dependent studies were conducted. For the time course study, fibroblasts were treated with BK (0.1 µM, Tocris, Minneapolis, MN, USA) or DABK (0.1 µM, MedChemExpress, Princeton, NJ, USA) for 1, 2, 4, 8, and 12 h. For the concentration-dependent study, fibroblasts were treated with BK or DABK for 4 h at concentrations of 0.01, 0.1, and 1 µM. To examine the receptor subtype involved, fibroblasts were treated with BK (0.1 µM) or DABK (0.1 µM) for 4 h in the presence or absence of the BDKRB1 antagonist ELN-441958 (1 µM, MedChemExpress) or the BDKRB2 antagonist icatibant (1 µM, MedChemExpress). The antagonists were added 1 h before BK or DABK treatment. To examine the downstream signaling pathway involved in the induction of PTGS2 expression by BK or DABK, a time course study (15, 30, 60, 120, and 180 min) was carried out to examine the effect of BK (0.1 µM) or DABK (0.1 µM) on the phosphorylation of MAPK members including p38, ERK1/2, and JNK. To further investigate the involvement of MAPK members in the induction of PTGS2 expression by BK or DABK, fibroblasts were treated with BK (0.1 µM) or DABK (0.1 µM) for 4 h in the presence or absence of the p38 inhibitor SB203580 (10 µM; Selleck, Houston, TX, USA), the ERK1/2 inhibitor PD98059 (20 µM; Selleck), or the JNK inhibitor SP600125 (10 µM; Sigma). The inhibitors were added 1 h before BK or DABK treatment.

To study the effects of SAA1 and LPS on BDKRB1 and BDKRB2 expression in amnion fibroblasts, cells were treated with recombinant human apo-SAA1 (10, 50, and 100 ng/ml; PeproTech Inc., Rocky Hill, NJ, USA) or LPS (1, 10, and 50 ng/ml; Sigma) for 24 h. The involvement of toll-like receptor 4 (TLR4) in the effects of SAA1 and LPS on BDKRB1 and BDKRB2 was studied by treating the cells with SAA1 (50 ng/ml) or LPS (50 ng/ml) in the presence or absence of the TLR4 receptor antagonist CLI-095 (5 µM, Invitrogen, San Diego, CA, USA) for 24 h. To examine the effect of SAA1 and LPS on the induction of PTGS2 expression by BK and DABK, the cells were pretreated with SAA1 (50 ng/ml) or LPS (50 ng/ml) for 24 h, and then the culture medium was replaced with medium containing only BK (0.1 µM) or DABK (0.1 µM) for further incubation for 4 h.

For the measurement of BDKRB1 and BDKRB2 mRNA abundance in human amnion tissue, the snap-frozen tissue was ground in liquid nitrogen and homogenized. Total RNA was extracted from the homogenized tissue with a total RNA isolation Kit (Foregene, Chengdu, China). For the measurement of BDKRB1, BDKRB2, and PTGS2 mRNA abundance in human amnion fibroblasts or epithelial cells, total RNA was extracted from the cells with the same total RNA isolation Kit. The mRNA in total RNA was reverse-transcribed to cDNA using a commercial kit (TaKaRa, Tokyo, Japan). The reverse-transcribed cDNA was used to measure BDKRB1, BDKRB2, and PTGS2 mRNA in power SYBR^® Premix Ex Taq™ (TaKaRa) with qRT-PCR. The housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was amplified in parallel as an internal control. The abundance of mRNA was quantified using the 2−^△△Ct method. The primer sequences used for qRT-PCR are illustrated in Table 3 .

After treatments, human amnion fibroblasts were lysed in ice-cold radioimmunoprecipitation assay (RIPA) buffer (Active Motif, Carlsbad, CA, USA) containing a protease inhibitor cocktail (Roche, Basel, Switzerland) and a phosphatase inhibitor (Roche) for the extraction of total cell protein for the measurements of PTGS2, MAPKs, and phosphorylated MAPKs with Western blotting. For the measurements of membrane BDKRB1 and BDKRB2 protein abundance, membrane protein was extracted with a differential centrifugation method using a commercial kit (Beyotime, Shanghai, China). After determination of protein concentration, the extracted protein was analyzed with Western blotting following a standard protocol. The primary antibodies used for Western blotting were as follows: p38 (1:500, Cell Signaling, Danvers, MA, USA, #8690), phosphorylated p38 at Thr180/Tyr182 (1:500, Cell Signaling, #4511), ERK1/2 (1:500, Cell Signaling, #4631), phosphorylated ERK1/2 at Thr202/Tyr204 (1:500, Cell Signaling, #4370), JNK (1:1,000, Cell Signaling, #9252), phosphorylated JNK at Thr183/Tyr185 (1:1,000, Cell Signaling, #9255), BDKRB1 (1:100, Santa Cruz, Santa Cruz, CA, USA, #sc-293196), BDKRB2 (1:500, Alomone, Jerusalem BioPark, Israel, #ABR-012), and PTGS2 (1:1,000, Cell Signaling, #12282). After incubation with primary antibodies for 24 h, the corresponding secondary antibodies conjugated with horseradish peroxidase were applied to further incubate for 1 h. Peroxidase activity was developed by a chemiluminescence detection system (Millipore, Billerica, MA, USA) and visualized using a G-Box chemiluminescence image capture system (Syngene, Cambridge, UK). Internal loading controls were probed with a GAPDH antibody (1:10,000, Proteintech, #60004-1-Ig) for total cell protein or Na⁺/K⁺ ATPase antibody (1:500, Cell Signaling, #23565) for membrane protein.

After treatment with BK (0.1 µM) or DABK (0.1 µM) for 4 h or pretreatment with SAA1 (50 ng/ml, 24 h) and LPS (50 ng/ml, 24 h) followed by BK (0.1 µM) or DABK (0.1 µM) treatment for 4 h, PGE2 in the culture media was measured with an enzyme immunoassay kit (Cayman, Ann Arbor, MI, USA) according to the protocol provided by the manufacturer.

All data are reported as means ± SEM. The number of each study represents separate experiments using amnion from different women. The Shapiro–Wilk normality test was used to examine the normal distribution of the data. Unpaired Student’s t-test was employed for unpaired data of two groups with normal distribution. The Wilcoxon signed-rank test and Mann–Whitney U test were used for paired data and unpaired data of two groups, respectively, when the data were not normally distributed. One-way ANOVA test followed by Tukey test or Dunnett’s test was performed to analyze paired data of more than two groups with normal distribution. For the analysis of SAA1 or LPS interaction with BK or DABK, the two-way ANOVA test was employed. Significance was set at p <0.05.

The transcriptomic sequencing data revealed that the transcripts of KNG1, KLKB1, and KLK1, the genes encoding kininogen, plasma, and tissue kallikrein, respectively, were hardly detectable in human amnion, suggesting that bradykinin peptides cannot be synthesized locally in human amnion. By contrast, BDKRB1 and BDKRB2 transcripts, the genes encoding BDKRB1 and BDKRB2, respectively, were detected in human amnion with more abundant BDKRB1 transcripts than BDKRB2 transcripts ( Figure 1B ). Moreover, the transcriptomic sequencing data showed that the abundance of both BDKRB1 and BDKRB2 transcripts was significantly increased in human amnion at TL as compared with TNL ( Figure 1B ). The transcriptomic sequencing data are available at NCBI GEO with the accession number GSE166453. Immunohistochemical staining showed that both BDKRB1 and BDKRB2 were distributed mainly in mesenchymal fibroblasts of human amnion although BDKRB2 appeared also in amnion epithelial cells ( Figure 1C ), which was confirmed by measuring BDKRB1 and BDKRB2 mRNA in amnion fibroblasts and epithelial cells with qRT-PCR ( Figure 1C ). These data suggest that amnion mesenchymal fibroblasts are an important target of bradykinin actions which may be enhanced at parturition.

Assay with qRT-PCR confirmed the increases in BDKRB1 and BDKRB2 mRNA abundance in human amnion at TL ( Figures 2A, B ). In addition, increases in BDKRB1 and BDKRB2 mRNA abundance were also observed in human amnion at PL ( Figures 2C, D ). Despite the absence of local bradykinin synthesis, bradykinin peptides were detected with ELISA in human amnion, which was also increased at both TL and PL ( Figures 2E, F ). These data suggest that bradykinin peptides detected in human amnion may be derived from adjacent tissues or maternal blood, and the actions of bradykinin peptides are enhanced in human amnion at both term and preterm labor.

Time course studies showed that both BK and DABK induced PTGS2 expression in human amnion fibroblasts in a time-dependent manner ( Figures 3A, B ). Significant increases in PTGS2 mRNA were observed with BK (0.1 μM) treatment for 1, 2, and 4 h but not for 8 and 12 h with maximal effect observed around 2 h. DABK (0.1 μM) treatment significantly increased PTGS2 mRNA abundance in amnion fibroblasts at 1, 2, 4, and 8 h with maximal effect observed between 1 and 4 h. The changes in PTGS2 protein lagged behind that of mRNA. Significant increases in PTGS2 protein were observed at 4 and 8 h after BK (0.1 μM) treatment and at 4, 8, and 12 h after DABK (0.1 μM) treatment. Since both PTGS2 mRNA and protein were significantly increased at 4 h, subsequent studies adopted the incubation with BK and DABK for 4 h.

The induction of PTGS2 mRNA and protein by BK and DABK in amnion fibroblasts was also concentration-dependent ( Figures 3C, D ). Significant increases in PTGS2 mRNA and protein abundance were observed at concentrations greater than 0.1 μM for both BK and DABK. Therefore, 0.1 μM of BK and DABK was chosen for the subsequent studies. At 0.1 μM, BK and DABK also significantly increased PGE2 concentration in the culture medium of amnion fibroblasts ( Figures 3E, F ).

The induction of PTGS2 mRNA and protein by BK (0.1 μM, 4 h) in amnion fibroblasts was blocked by the BDKRB2 antagonist icatibant (1 μM) but not by the BDKRB1 antagonist ELN-441958 (1 μM) ( Figures 4A, B ). By contrast, the induction of PTGS2 expression by DABK (0.1 μM, 4 h) was blocked by the BDKRB1 antagonist ELN-441958 (1 μM) but not by the BDKRB2 antagonist icatibant (1 μM) ( Figures 4C, D ). Both BK (0.1 μM) ( Figures 5A–C ) and DABK (0.1 μM) ( Figures 5D–F ) increased ERK1/2, p38, and JNK phosphorylation significantly in amnion fibroblasts at 15 min after treatments. Inhibition of ERK1/2 and p38 with PD98059 (20 μM) and SB203580 (10 μM), respectively, attenuated the induction of PTGS2 mRNA and protein by BK (0.1 μM) ( Figures 6A, B ) or DABK (0.1 μM) ( Figures 6C, D ) significantly. However, inhibition of JNK with SP600125 (10 µM) failed to block the induction of PTGS2 mRNA and protein by either BK (0.1 μM) or DABK (0.1 μM) ( Figures 6E, F ).

Both LPS and SAA1 increased BDKRB1 and BDKRB2 mRNA abundance in a concentration-dependent manner in amnion fibroblasts. LPS (24 h) significantly increased BDKRB1 mRNA at 50 ng/ml and increased BDKRB2 mRNA at 1, 10, and 50 ng/ml ( Figure 7A ). SAA1 (24 h) significantly increased BDKRB1 mRNA at 10, 50, and 100 ng/ml and BDKRB2 mRNA at 50 and 100 ng/ml ( Figure 7B ). Consistently, the protein abundance of BDKRB1 and BDKRB2 was also significantly increased by LPS (50 ng/ml, 24 h) and SAA1 (50 ng/ml, 24 h) ( Figures 7C, D ). The induction of BDKRB1 and BDKRB2 mRNA by LPS (50 ng/ml, 24 h) ( Figure 7E ) and SAA1 (50 ng/ml, 24 h) ( Figure 7F ) was blocked by the TLR4 antagonist CLI-095 (5 μM).

Pretreatment of amnion fibroblasts with LPS (50 ng/ml) or SAA1 (50 ng/ml) for 24 h significantly enhanced the subsequent induction of PTGS2 mRNA ( Figures 8A–E ) and protein expression ( Figures 8F–I ) and PGE2 production ( Figures 8J–M ) by BK (0.1 μM, 4 h) or DABK (0.1 μM, 4 h).