Résumé
Objective To examine the effects of Toxoplasma gondii infection on the proportion, quantity, differentiation and function of mouse and human uterine natural killer cells (uNK cells), so as to explore the role of uNK cells in abortion of early pregnancy caused by T. gondii infection. Methods Pregnant mice were injected intraperitoneally with T. gondii tachyzoites on day 6.5 of pregnancy, and the abortion mouse model caused by T. gondii infections was constructed. Mouse uterine lymphocytes were isolated on day 9.5 of pregnancy. Human uterine lymphocytes were isolated from fresh human decidual specimens after abortion in normal early pregnancy and co-cultured with tachyzoites of the T. gondii RH strain for 48 h at T. gondii/uterine lymphocytes ratios of 0.5:1, 1:1 and 2:1. The phenotypes of mouse uNK cells (CD122, NK1.1, DX5) and human uNK cells (CD3, CD56, CD11b, CD27) and the expression of intracellular cytokines interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) were detected by flow cytometry. Mouse and human uNK cells were sorted by magnetic beads, and the cytotoxicity of uNK cells was tested using the lactate dehydrogenase (LDH) release assay at effector/target cell ratios of 1:1, 5:1, 10:1 and 20:1 with mouse or human uNK cells as effector cells and mouse YAC-1 cells or human K562 cells as target cells. Results On day 9.5 of pregnancy, the mouse abortion rate was significantly higher in the infected group than that in the control group (83.02% vs. 3.51%; χ2 = 71.359, P < 0.001). Significantly lower absolute number of uNK cells [(4 547 ± 1 610) cells/mouse vs. (8 978 ± 3 339) cells/mouse; U = 2.000, P < 0.05], lower NK1.1 expression on uNK cell surface [(74.53 ± 8.37)% vs. (93.00 ± 1.11)%; U = 0.000, P < 0.05], higher proportion of NK1.1-DX5-cells [(20.10 ± 8.03)% vs. (5.04 ± 0.68)%; U = 0.000, P < 0.05], lower proportion of NK1.1+ DX5+ cells [(21.70 ± 12.48)% vs. (45.75 ± 2.26)%; U = 0.000, P < 0.05] and higher IFN-γ expression [(16.74 ± 1.36)% vs. (8.13 ± 1.90)%; U = 0.000, P < 0.05] were detected in the infected group than in the control group, while no significant difference was seen in TNF-α expression between the two groups [(67.98 ± 9.20)% vs. (52.93 ± 10.42)%; U = 2.000, P > 0.05]. The mouse uNK cells showed a strong cytotoxicity in the infected group, and the cytotoxicity gradually increased with the effector/target cell ratio. The cytotoxicity of uNK cells against YAC-1 cells was 2.30%, 4.32%, 8.12% and 12.65% in the infected group and 1.21%, 1.63%, 2.51% and 3.22% in the control group at effector/target cell ratios of 1:1, 5:1, 10:1 and 20:1, respectively. Following co-culture of human uterine lymphocytes and tachyzoites of the T. gondii RH strain for 48 h, the proportion [TOX 2:1 group vs. control group: (6.61 ± 1.75)% vs. (17.48 ± 4.81)%; F = 7.307, P < 0.01], and absolute number of human uNK cells in uterine lymphocytes of human uNK cells in uterine lymphocytes [TOX 2:1 group vs. control group: (12 104 ± 5 726) cells/well vs. (65 285 ± 21 810) cells/well; H = 11.540, P < 0.01] were significantly lower in the infected group than in the control group. A lower proportion of CD56brightCD16- NK cells [TOX 2:1 group vs. control group: (25.25 ± 5.90)% vs. (36.03 ± 4.51)%; F = 3.213, P > 0.05] and higher proportion of CD56dimCD16+ NK cells [TOX 2:1 group vs. control group: (11.15 ± 2.15)% vs. (7.09 ± 2.24)%; F = 2.992, P > 0.05] were detected in uNK cells in the infected group than in the control group, and the ratio of CD56brightCD16- cells/CD56dimCD16+ cells was significantly lower in the infected group than in the control group [TOX2:1 group vs. control group: (2.37 ± 0.92) vs. (5.58 ± 2.39); H = 8.228, P < 0.05]. In addition, the proportion of CD11b+CD27- cells in human uNK cells was significantly higher in the infected group than in the control group [TOX 2:1 group vs. control group: (30.28 ± 6.91)% vs. (17.48 ± 4.67)%; H = 6.556, P < 0.05], while no significant differences were found between the two groups in terms of IFN-γ [TOX 2:1 group vs. control group: (14.13 ± 1.28)% vs. (15.19 ± 1.64)%; F = 1.639, P > 0.05] or TNF-α expression [TOX 2:1 group vs. control group: (54.76 ± 10.02)% vs. (50.33 ± 3.67)%; F = 0.415, P > 0.05]. Human uNK cells presented a strong cytotoxicity in the infected group, and the cytotoxicity gradually increased with the effector/target cell ratio. The cytotoxicity of human uNK cells against K562 cells was 11.90%, 28.11%, 49.91% and 73.35% in the infected group and 12.21%, 21.63%, 33.51% and 48.22% in the control group at effector/target cell ratios of 1:1, 5:1, 10:1 and 20:1, respectively. Conclusions T. gondii infection presents diverse effects on the differentiation and secretion ability of mouse and human uNK cells. However, T. gondii infection causes a reduction in the absolute number and enhances the cytotoxicity of both mouse and human uNK cells.
Résumé
Objective To investigate the effect of Zidan decoction containing serum on endometrial receptivity in uterine immunity environment,and then to discuss if it's effect on endometrial receptivity through its controlling the expression of cell factors which related to endometrial receptivity.Methods Mice's endometrium tissue samples and uterine Natural Killer (UNK) cells were obtained from pregnant mice on gd 4.5 days and gd 12.5 days.After endometrial glandular epithelial cells,endometrial stromal cells and UNK cells were separated through primary culture,epithelial cells and endometrial stromal cells at rate of 1 ∶ 1 a certain concentration UNK were mixed together and highly purified-cultured.Different concentrations of Zidan decoction was acted on the culture solution.FBS,black and aspirin control group were setup for contrast.The expression of LIF and VEGF which influenced by Zidan decoction in the immunity environment was assayed by Realtime-PCR.Results The expression of LIF and VEGF was significantly increased in the immunity environment of uterine (in 10％ Zidan decoction group,10％ aspirin group and 10％ blank control group:LIF was 2.10±0.20,1.98±0.14,0.90±0.05 separately,and VEGF was 3.17±0.31,2.52±0.09,0.92±0.06respectively,P＜0.05).Conclusion Zidan decoction could improve endometrial receptivity by increasing the expression of cell factor which related to endometrial receptivity with the help ofUNK paracrine.
Résumé
Mouse and vole embryos were allogeneically and xenogeneically transferred into pseudopregnant CD.1 and immunodeficient (seid)female mice,and we investigated the distribution of uterine leucocytes cells in the implantation sites on days 5,6,and 7 of pregnancy. Maerophages were evenly distributed in the endometrium on days 5-7.Neutrophils were rarely seen on days 5-7,but lymphocytes were found throughout the endometrium,often in groups associated with glands or the luminal epithelium.The number of uNK cells increased markedly at the mesometrial uriangle and the outer decidual area in the CD-1 uteri containing vole embryos;by contrast,seid uteri having vole embryos showed almost the same number as those having mouse embryos.Mast cells were present in large numbers at the myometrium,but rarely in the decidua in all types of pregnant uteri.Cells at the myometrium were more numerous in xenogeneic than in allogeneic transfer.Maay mast cells appeared in the inner decidua where xenogeneically transferred vole embryos were dead and aborted.These results suggest the possibility that uterine leucocytes mediate various immunological events in the mouse-vole interspesific pregnancies.