Peritoneal Fluid from Patients with Ovarian Endometriosis Displays Immunosuppressive Potential and Stimulates Th2 Response

Abstract

Endometriosis is a common gynaecological disorder characterized by the ectopic growth of endometrial tissue outside the uterine cavity. It is associated with chronic pelvic inflammation and autoimmune reactivity manifesting by autoantibody production and abrogated cellular immune responses. Endometriotic peritoneal fluid contains various infiltrating leucocyte populations and a bulk of proinflammatory and immunoregulatory cytokines. However, the nature and significance of the peritoneal milieu in women with endometriosis still remains obscure. Therefore, the aim of the present study was to investigate the immunoregulatory activity of the peritoneal fluid (PF) from women with endometriosis. The peritoneal fluid samples were collected during laparoscopic surgery from 30 women with and without endometriosis. Immunoregulatory cytokines (IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ and TNF) and chemokines (CCL2, CCL5, CXCL8 and CXCL9) were evaluated in PF and culture supernatants generated by unstimulated and CD3/CD28/IL-2-stimulated CD4⁺ T cells cultured in the presence of PF. The effect of PF on the generation of Treg and Th17 cells in CD4⁺ T cell cultures, as well as the natural cytotoxic activity of peripheral blood mononuclear cells, was also investigated. Concentrations of IL-6, IL-10, CCL2, CXCL8 and CXCL9 were significantly upregulated in the PF from women with endometriosis when compared to control women, whereas concentrations of other cytokines and chemokines were unaffected. The culturing of unstimulated and CD3/CD28/IL-2-stimulated CD4⁺ T cells in the presence of endometriotic PF resulted in the downregulation of their IL-2, IFN-γ, IL-17A and TNF production as compared to culture medium alone. On the other side, endometriotic PF significantly stimulated the production of IL-4 and IL-10. Endometriotic PF also stimulated the release of CCL2 and CXCL8, whereas the production of CCL5 and CXCL9 was downregulated. Endometriotic PF stimulated the generation of Treg cells and had an inhibitory effect on the generation of Th17 cells in cultures of CD4⁺ T cells. It also inhibited the NK cell cytotoxic activity of the peripheral blood lymphocytes. These results strongly imply that the PF from patients with endometriosis has immunoregulatory/immunosuppressive activity and shifts the Th1/Th2 cytokine balance toward the Th2 response, which may account for deviation of local and systemic immune responses. However, a similar trend, albeit not a statistically significant one, was also observed in case of PF from women without endometriosis, thus suggesting that peritoneal milieu may in general display some immunoregulatory/immunosuppressive properties. It should be stressed, however, that our present observations were made on a relatively small number of PF samples and further studies are needed to reveal possible mechanism(s) responsible for this phenomenon.

Keywords: Th1 cells; Th17 cells; Th2 cells; Treg cells; chemokines; cytokines; endometriosis; peritoneal fluid.

Figure 1

Production of (A) IL-2, (B) IFN-γ, (C) IL-17A, (D) TNF, (E) IL-4, (F) IL-10 and (G) IL-6 by cultured CD4+ T cells unstimulated or stimulated with CD3/CD28 beads and IL-2 in the presence of the culture medium alone (CM), peritoneal fluid from control woman (Control PF) or peritoneal fluid from woman with endometriosis (ENDO PF). The results are shown as scatter dot plots with a median and interquartile range. Statistical significance was computed by paired non-parametric ANOVA (Friedman’s test) followed by a post hoc test. * Statistically significant from the control group at least at p < 0.05. Baseline concentration ranges of the tested cytokines (pg/mL), respectively, in Control PF and ENDO PF used for the experiments were as follows. IL-2, 0.13–0.56 and 0.13–0.88; IFN-γ, 0–0.82 and 0–0.95; IL-17A, 0–8.10 and 0–8.70; TNF, 0.43–1.32 and 0.38–2.24; IL-4, 0–0.70 and 0–0.59; IL-10, 0.59–9.65 and 1.46–11.2; IL-6, 16.4–312.8 and 47.1–492.

Figure 2

Production of (A) CCL2, (B) CCL5, (C) CXCL8 and (D) CXC9 by cultured CD4⁺ T cells unstimulated or stimulated with CD3/CD28 beads and IL-2 in the presence of the culture medium alone (CM), peritoneal fluid from control woman (Control PF) or peritoneal fluid from woman with endometriosis (ENDO PF). The results are shown as scatter dot plots with a median and interquartile range. Statistical significance was computed by paired non-parametric ANOVA (Friedman’s test) followed by a post hoc test. * Statistically significant from the control group at least at p < 0.05. Baseline concentration ranges of the tested chemokines (pg/mL), respectively, in Control PF and ENDO PF used for the experiments were as follows. CCL2, 16.6–198.7 and 10.12–393.2; CCL5, 2.5–9.6 and 6.0–33.9; CXCL8, 7.08–61.98 and 15.4–400.0; CXC9, 12.1–50.8 and 39.2–72.6.

Figure 3

Effect of culture medium alone (CM), peritoneal fluid from control woman (Control PF) or peritoneal fluid from woman with endometriosis (ENDO PF) on generation of CD25^high and CD25^high FOXP3⁺ Treg cells in cultures of CD4⁺ T cells unstimulated or stimulated with CD3/CD28 beads and IL-2. (A) Gating strategy and a representative flow cytometry analysis showing identification of the respective CD25^high and CD25^high FOXP3⁺ T cell subpopulations in CD4⁺ T cells under different culture conditions. (B) Proportions of CD25^high T cells and (C) CD25^high FOXP3⁺ Treg cells in population of unstimulated or CD3/CD28 beads+IL-2-stimulated CD4⁺ T cells. The results are shown as scatter dot plots with a median and interquartile range. Statistical significance was computed by paired (Friedman’s test) or unpaired (Kruskal–Wallis test) non-parametric ANOVA followed by a post hoc test.

Figure 4

Effect of culture medium alone (CM), peritoneal fluid from control woman (Control PF) or peritoneal fluid from woman with endometriosis (ENDO PF) on generation of CD161⁺ and CD161⁺ RORγ⁺ Th17 cells in cultures of CD4⁺ T cells unstimulated or stimulated with CD3/CD28 beads and IL-2. (A) Gating strategy and a representative flow cytometry analysis showing identification of the respective CD161⁺ and CD161⁺ RORγ⁺ T cell subpopulations in CD4⁺ T cells under different culture conditions. (B) Proportions of CD161⁺ T cells and (C) CD161⁺ RORγ⁺ Th17 cells in population of unstimulated or CD3/CD28 beads+IL-2-stimulated CD4⁺ T cells. The results are shown as scatter dot plots with a median and interquartile range. Statistical significance was computed by paired (Friedman’s test) or unpaired (Kruskal–Wallis test) non-parametric ANOVA followed by a post hoc test.

Figure 5

Effect of culture medium alone (CM), peritoneal fluid from control woman (Control PF) or peritoneal fluid from woman with endometriosis (ENDO PF) on the NK cell cytotoxic activity of cultured PBMC. (A) The representative flow cytometry analysis showing the controls for the NK assay. Shown are the fluorescence-negative effector cells (PBMC), green fluorescence (FITC) labeled K562 target cells and spontaneously dying K562 cells (red fluorescence, PE). Spontaneous death of target cells was determined in cultures without effector cells. (B) An example of identification of target K562 cells killed by NK cells from the PBMC population (red fluorescence). (C) Relative cell mediated cytotoxicity of untreated and peritoneal fluid-treated PBMC against K562 cells. The results are expressed as an index of specific cytotoxicity of peritoneal fluid-treated PBMC relative to untreated control PBMC. Each bar represents mean ± SD from 4 independent experiments. Statistical significance was computed by paired non-parametric ANOVA (Friedman’s test) followed by a post hoc test.