Anti-inflammatory effects of progesterone through NF-κB and MAPK pathway in lipopolysaccharide- or Escherichia coli-stimulated bovine endometrial stromal cells.

After parturition, the bovine uterine lumen becomes contaminated, and persistence of pathogenic bacteria often leads to uterine infection, such as metritis and endometritis. These uterine impairments are associated with poor reproductive performance even after successful resolution of the disease, and eventually cause huge financial losses [ 1 ]. Among various bacteria that contaminate bovine uterus, Escherichia coli ( E . coli ) are abundant and are the first step in the pathogenesis of postpartum uterine disease in cattle [ 2 ]. Classical innate immunity is a principal component of uterine defense and encompasses anatomical, physiological, phagocytic, and inflammatory barriers [ 1 ]. The endometrial epithelial barrier is disrupted due to parturition, making the underlying stroma exposed to pathogenic microorganisms. Lipopolysaccharide (LPS) is the main endotoxin of E . coli . Exposure of bovine endometrial cells with LPS or E . coli results in inflammatory response through pattern recognition receptors [ 2 ]. Toll-like receptor 4 (TLR4) binds to LPS together with myeloid differentiation factor 2 and CD14 on endometrial cell surface, followed by the activation of nuclear factor-κB (NF-κB) transcription factors and the mitogen-activated protein kinases (MAPK) extracellular signal, including extracellular regulated kinase (ERK), p38, and c-Jun N-terminal kinase (JNK), and finally results in an enhanced expression of transcripts for cytokines and chemokines such as interleukin 1 beta (IL1β), IL6, tumor necrosis factor-α (TNF-α), IL8, nitric oxide, and prostaglandins [ 2 – 4 ]. Progesterone has been proved to show immunosuppressive function within the endometrium of ruminants, and the pathogenesis of postpartum uterine infection can be related to progesterone [ 5 ]. Normally after parturition, progesterone concentration keeps low or undetectable until after the first ovulation [ 6 – 8 ]. But an early ovulation could make the cow at a higher risk of uterine infection if uterine involution is incomplete [ 9 ]. Spontaneous uterine infection in cattle do not usually develop until after the first postpartum corpus luteum forms and begins producing progesterone [ 10 ]. In practice, we found the presence of corpus luteum in about 25% of cows diagnosed with endometritis. Mechanism underlying the immunosuppressive role of progesterone on uterus of ruminant species was associated with decreased of lymphocyte proliferation, modulation on amount and function of polymorphonuclear leukocyte, and regulation of inflammatory cytokines and chemokines [ 10 – 13 ]. Generally, progesterone exhibited an anti-inflammatory property in various cell types [ 14 – 16 ], and has been shown to perturbate prostaglandin secretions in bovine endometrial epithelial and stromal cells in response to LPS or E . coli challenge [ 3 ]. But the study by Saut et al. found no impact of progesterone on the inflammatory response of bovine endometrial epithelial and stromal cells stimulated by LPS [ 4 ]. Previously we verified that progesterone inhibited LPS- or E . coli -induced expression of proinflammatory genes and MAPK and NF-κB pathways in primary bovine endometrial epithelial cells [ 17 ]. Compared with epithelial cells, stromal cells are more susceptible to pathogenic E . coli with a stronger binding affinity and a more intense inflammatory response [ 18 ]. However, the effect of progesterone on bovine endometrial stromal cells (BESC) remains controversial. In the present study, we hypothesized that progesterone ameliorates inflammatory response in bovine endometrial stromal cells. Primary BESC was stimulated with LPS or heat-killed E . coli , and was treated with physiological concentrations of progesterone (1, 3, or 5 ng/mL). The changes in mRNA expressions of IL1B , IL6 , TNF , C-X-C motif chemokine ligand 8 ( CXCL8 ), nitric oxide synthase 2 ( NOS2 ), and prostaglandin-endoperoxide synthase 2 ( PTGS2 ), and activations of NF-κB and MAPK pathways were determined.

The BESC viability was not influenced ( P > 0.05) by the treatment of progesterone, LPS, or heat-killed E . coli individually or in combination ( Fig 1 ). The cells were treated with 1, 3, or 5 ng/mL progesterone (A), 1 μg/mL lipopolysaccharide and progesterone (B), or 1×10 8 CFU/mL heat-killed E . coli and progesterone (C) for 24 h. The vehicle is 0.5% alcohol. P4, progesterone. LPS, lipopolysaccharide. All data were presented as means ± SEM (n ≥ 3). The impact of progesterone alone on proinflammatory gene expressions was evaluated at 12 h. This time point was selected according to a pre-experimental result (unpublished data, S1 Appendix ). The mRNA expressions of IL6 , TNF , CXCL8 , NOS2 , or PTGS2 were unaffected ( P > 0.05) by 1, 3, or 5 ng/mL progesterone. However, progesterone decreased ( P < 0.05) IL1B mRNA expression ( Fig 2A ). Based on a pilot study, time points of 12 and 24 h exhibited peak abundance of mRNA transcripts for proinflammatory genes in LPS-treated BESC (unpublished data, S1 Appendix ). As shown in Fig 2B , exposure of cells to LPS upregulated ( P < 0.05) relative mRNA abundance of IL1B , IL6 , TNF , CXCL8 , NOS2 , and PTGS2 . Compared with LPS group, progesterone (1, 3, and 5 ng/mL) decreased ( P < 0.05) mRNA expressions of IL1B , IL6 , CXCL8 , NOS2 , and PTGS2 in response to LPS at 12 and 24 h. Down-regulation ( P < 0.05) of TNF mRNA expression was only observed in cells cotreated with LPS and 5 ng/mL progesterone at 12 h, and in cells cotreated with LPS and progesterone (1 or 3 ng/mL) at 24 h. The time course experiment showed that heat-killed E . coli induced expression of proinflammatory genes ( P < 0.05), and the increase was most prominent at 18 and 24 h ( Fig 2C ). Progesterone (1, 3, and 5 ng/mL) generally reduced ( P 0.05) TNF and NOS2 at 24 h in cells cotreated with E . coli and 5 ng/mL P4 ( Fig 2D ). The mRNA expressions of IL1B , IL6 , CXCL8 , TNF , NOS2 , and PTGS2 were assessed by qPCR (normalized to ACTB ). A. Unstimulated cells were treated with 1, 3, and 5 ng/mL progesterone for 12 h. B. Cells were cotreated with 1 μg/mL lipopolysaccharide and progesterone (1, 3, and 5 ng/mL) for 12 and 24 h. C. Time-course changes in inflammatory gene expressions in cells treated with 1×10 8 CFU/mL heat-killed E . coli for 6, 9, 12, 18, and 24 h. D. Cells were cotreated with 1×10 8 CFU/mL heat-killed E . coli and progesterone (1, 3, and 5 ng/mL) for 18 and 24 h. LPS, lipopolysaccharide. All data were presented as means ± SEM (n ≥ 3). * P < 0.05, difference compared with the control; # P 0.05) key protein levels in NF-κB pathway. Our preliminary study showed that the phosphorylations of IκBα and P65 elevated at 45 min in response to LPS stimulation (unpublished data, S1 Appendix ). The addition of progesterone decreased ( P < 0.05) LPS-induced phosphorylations of P65 and IκBα. The heat-killed E . coli resulted in phosphorylations of IκBα and P65 from 30 to 120 min, and was maximal ( P < 0.05) at 120 min ( Fig 3B ). The amount of phosphorylations of IκBα and P65 in cells cotreated with progesterone (1, 3, or 5 ng/mL) and heat-killed E . coli was generally less ( P < 0.05) than in cells treated only with heat-killed E . coli . The protein level of IκBα, P-IκBα, P65, and P-P65 was determined using Western blot. The protein blots were quantified by Quantity One software. A. Unstimulated cells were treated with 1, 3, and 5 ng/mL progesterone for 30 min. B. The cells were cotreated with 1 μg/mL lipopolysaccharide and progesterone (1, 3, or 5 ng/mL) for 45 min. This time point was based on the results of our previous study. C. Time-course changes in key protein phosphorylation of NF-κB pathway in cells challenged with 1×10 8 CFU/mL heat-killed E . coli for 30, 60, 90, and 120 min. D. The cells were cotreated with heat-killed E . coli and progesterone for 120 min. LPS, lipopolysaccharide. P4, progesterone. All data were presented as means ± SEM (n ≥ 3). * P < 0.05, difference compared with the control; # P < 0.05, difference compared with the LPS group (B) or the heat-killed E . coli group (D). The immunofluorescence results showed translocation of NF-κB P65 into the nucleus in BESC stimulated with LPS and heat-killed E . coli ( Fig 4 ). The addition of 3 ng/mL progesterone reduced the amount of P65 in nucleus in cells challenged with LPS or bacteria. The cells were cotreated with 1 μg/mL lipopolysaccharide and 3 ng/mL progesterone for 45 min (A), or cotreated with 1×10 8 CFU/mL heat-killed E . coli and 3 ng/mL progesterone for 120 min (B). LPS, lipopolysaccharide. P4, progesterone. The cells treated with only progesterone (1, 3, and 5 ng/mL) showed no change ( P > 0.05) in key protein levels of MAPK pathway ( Fig 5A ). Treatment of BESC with 1 μg/mL LPS induced phosphorylations of ERK, JNK, and P38 at 60 min (data unpublished, S1 Appendix ). As shown in Fig 5B , compared with LPS group, the ratios of P-ERK/ERK, P-JNK/JNK, and P-P38/P38 were generally lower in cells of cotreatment groups, except the P-JNK/JNK ratio in cells cotreated with LPS and progesterone (1 or 3 ng/mL). Heat-killed E . coli caused elevation ( P < 0.05) in the phosphorylations of ERK, JNK, and P38, with peak levels at 120 min ( Fig 5C ). Similarly, progesterone (1, 3, and 5 ng/mL) treatment reduced ( P < 0.05) the phosphorylations of ERK, JNK, and P38 in response to bacterial challenge ( Fig 5D ). The protein levels of ERK, P-ERK, JNK, P-JNK, P38, and P-P38 was determined using Western blot. The blots were quantified by Quantity One software. A. Unstimulated cells were treated with 1, 3, and 5 ng/mL progesterone for 30 min. B. The cells were cotreated with 1 μg/mL lipopolysaccharide and progesterone (1, 3, or 5 ng/mL) for 60 min. The time point was based on the results of our previous study. C. Time-course changes in key protein phosphorylation of MAPK pathway in cells challenged with 1×10 8 CFU/mL heat-killed E . coli for 30, 60, 90, and 120 min. D. The cells were cotreated with heat-killed E . coli and progesterone for 120 min. LPS, lipopolysaccharide. P4, progesterone. All data were presented as means ± SEM (n ≥ 3). * P < 0.05, difference compared with the control; # P < 0.05, difference compared with the LPS group (B) or the heat-killed E . coli group (D).

Progesterone elicited anti-inflammatory effect in LPS- or heat-killed E . coli -stimulated primary bovine endometrial stromal cells by downregulating gene expressions of IL1B , IL6 , TNF , CXCL8 , NOS2 , and PTGS2 corresponding to inhibitions of NF-κB and MAPK activation. Except the downregulation of basal IL1B transcription, progesterone alone showed no influence on these proinflammatory gene expressions and NF-κB and MAPK signaling cascades in unstimulated BESC.

All experimental procedures were approved by the Animal Care and Use Committee of Yangzhou University (NSFC2020-SYXY24). The bovine uteri without gross evidence of genital disease or infection were collected aseptically at local abattoir and were kept on ice until further processing in the laboratory. The uterine surface was disinfected with iodophor and 75% alcohol and was flushed clean with sterile saline. The uterine horn was dissected longitudinally to expose the endometrium. The epithelium was removed by using a moisturized cotton ball. Then the intercaruncular endometrial stripes were dissected from the myometrial layer with a scalpel blade and were chopped into small pieces. After repeated rinses with phosphate-buffered saline (PBS) supplemented with 100 U/mL penicillin/streptomycin, the minced tissue was digested in DMEM/F12 (D8900, sigma, USA) containing 0.25% collagenase II (C6885-5G, Sigma, USA) at 37°C for 50 min. The cell suspension was filtered through a 450 μm mesh to remove undigested tissue fragments, and the filtrate was washed 3 times by centrifugation (100× g for 5 min) with PBS. The cells were resuspended in DMEM/F12 containing 15% fetal bovine serum and 100 U/mL penicillin/streptomycin and were cultured at 37°C with 5% CO 2 . The medium was changed 12 h after plating to allow attachment of stromal cells and removal of epithelial cells. The purity of stromal cell population was determined to be more than 95% by the detection of vimentin using immunocytochemistry. The culture media were changed every 24 to 48 h once confluence had been reached approximately 90%. Both LPS and heat-killed E . coli were used to induce the inflammatory response in BESC. LPS (L2630) and progesterone (P0130) were purchased from Sigma-Aldrich. LPS was dissolved in DMEM/F12 at a concentration of 1 mg/mL as a stock solution at -20°C, and was diluted to 1 μg/mL with DMEM/F12 during experiment. Progesterone was dissolved and diluted in ethanol to 10 4 ng/mL as a stock solution at -20°C, and was diluted to 1, 3, or 5 ng/mL with DMEM/F12 as working solutions [ 17 ]. The heat-killed E . coli O55:B5 was prepared as previously described [ 19 ]. The inactive bacteria were resuspended in DMEM/F12 to a final inoculum of 1×10 8 CFU/mL. To determine the effect of progesterone on the inflammatory response of BESC, the cells were plated in 6-, 24-, or 96-well plates or 6-cm dishes and were challenged with 1 μg/mL LPS or 1×10 8 CFU/mL heat-killed E . coli in control medium or medium containing progesterone. One, 3, and 5 ng/mL progesterone were selected because the serum concentration of progesterone normally ranges from 1 to 5 ng/mL in cattle [ 4 ]. Based on our previous reports in epithelial cells and a pilot study in BESC, LPS of 1 μg/mL or heat-killed E . coli of 1×10 8 CFU/mL was able to induce the inflammatory response [ 19 ]. The cell viability was determined 24 h after treatment. The relative mRNA abundance of inflammatory genes was detected 12 and 24 h after the cotreatment of LPS and progesterone, or 18 and 24 h after the cotreatment of heat-killed E . coli and progesterone. The changes in key protein levels of NF-κB and MAPK pathways were detected at 45 and 60 min, respectively, in cells with cotreated with LPS and progesterone. The time point for the detection of NF-κB and MAPK pathways was 120 min in cells cotreated with heat-killed E . coli and progesterone. P65 nuclear translocation was detected using immunofluorescence at 45 or 120 min in cells stimulated with LPS or bacteria, respectively. In addition, to observe the impact of progesterone in unstimulated conditions, BESC was treated with progesterone alone for 12 h to detect proinflammatory gene expressions, and for 30 min to detect phosphorylation of key proteins in MAPK and NF-κB pathways. The selection of these time points was based on pre-experimental results. The Cell Counting Kit-8 (CCK-8, Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was used to evaluate the impact of progesterone, LPS, and heat-killed E . coli , either alone or in combination, on the cell viability of BESC. The cells were seeded into wells of a 96-well plate (2×10 3 cells per well) and grown to 80% confluence. The medium was replaced with DMEM/F12 containing progesterone, LPS and progesterone, or heat-killed E . coli and progesterone. After 24 h treatment, the CCK-8 solution was added to each well, followed by an additional incubation for 2 h. The optical density was read at 450 nm using a microplate reader (Tecan, Austria). The cells were plated in 6-well plates (2×10 5 cells per well) and grown to 80% fusion. After the treatment as previously described, the cells were washed with PBS, and the total RNA was subsequently extracted using a Trizol reagent (ET111, TRAN, China) according to the manufacturer’s protocol. The extracted RNA was quantified using a Nanodrop 2000 spectrophotometer (Thermo, USA). The absorption ratio (A260/280) was determined to be between 1.8 and 2.0. The RNA was reverse transcribed into cDNA using the PrimerScript RT regent Kit gDNA Eraser (DRR047A, TaKaRa, Japan). The quantitative PCR was carried out using a CFX 96 Real-Time PCR Detection System (Bio-Rad, USA) as previously described [ 20 ]. The 2 -△△Ct method was used to analyze the relative mRNA abundance. The actin beta ( ACTB ) was used as an internal control. A single product was amplified by each primer pair. The products were purified and sequenced (TsingKe Biotech, Beijing, China), and the sequence results were analyzed using BLAST ( http://blast.ncbi.nlm.nih.gov/blast.cg ) and compared to GenBank database. The sequences of primers were shown in Table 1 . The cells were seeded in 6-cm culture dishes (6×10 5 cells per dish) and grown to 80% confluence. After treatment, the cells were collected and lysed using a RIPA buffer (P0013B, Beyotime, China) containing protease and phosphatase inhibitor cocktail (P1046, Beyotime, China) for routine protein extraction. The total proteins were quantified using a bicinchoninic acid protein assay kit (P0010, Beyotime, China). Cell protein extract (20 to 30 μg) was subjected to a 10% SDS-polyacrylamide gel and was transferred to a polyvinylidene difluoride membrane (Millipore, Germany). The samples were blocked with Tris-buffered saline containing 0.05% Tween 20 and 5% skimmed milk, followed by the incubation with a primary antibody (1:1000 dilution with 5% bovine serum albumin) at 4°C overnight. The primary antibodies specific for β-actin (# 4970), IκBα (# 4812), P-IκBα (# 2859), P65 (# 8242), P-P65 (# 3033), ERK1/2 (# 4695), P-ERK 1/2 (# 4370), JNK (# 9258), P-JNK (# 4668), P38 (# 8690), and P-P38 (# 4511) were purchased from Cell Signaling Technology. The membranes were subsequently incubated with an HRP-conjugated secondary antibody (111-035-003, Jackson ImmunoResearch, USA; 1:10000 dilution with 5% skimmed milk) at room temperature for 1 h. The protein blots were detected and visualized using a chemiluminescence HRP substrate (1810202, Clinx Science Instruments, China) and a ChemiScope 5300 Pro CCD camera (Clinx Science Instruments, China). The blots were quantified by Quantity One software (Bio-Rad, CA, USA). The cells were plated in a 24-well culture plate and were treated according to the experiment design. Then the cells were fixed with 4% paraformaldehyde at room temperature for 12 min. After washing with PBS, the cell membrane was penetrated with 0.4% Triton X-100 (ST797, Beyotime, China) for 15 min. The cells were washed with PBS and were blocked using PBS containing 10% goat serum (ZLI-9021, ZSGB-BIO, China) for 45 min at room temperature. The cells were incubated with a primary antibody specific for NF-κB P65 (#8242, Cell Signaling Technology, USA) with 1:400 dilution at 4°C overnight, and were subsequently incubated with an FITC-conjugated secondary antibody (A11034, Thermo Fisher Scientific, USA) with 1:500 dilution for 1 h at room temperature in dark environment. The nuclei were stained with DAPI (C1005, Beyotime, China). The cells were visualized using a fluorescence microscope (Leica TCS Sp8, Leica company, Germany). All data were presented as the means ± standard error of means (SEM), and were analyzed using the SPSS 26.0 software (IBM, NY, USA). Statistically significant differences were calculated by one-way ANOVA, followed by Least Significant Difference test. A two-sided P -value less than 0.05 was designated as significant. Each experiment was repeated at least three times.

(DOCX) Click here for additional data file. (XLSX) Click here for additional data file. (ZIP) Click here for additional data file.