Interleukin (IL)-6: A Friend or Foe of Pregnancy and Parturition? Evidence From Functional Studies in Fetal Membrane Cells

Abstract

Objective: Protection of the fetus within the amniotic sac is primarily attained by remodeling fetal membrane (amniochorion) cells through cyclic epithelial to mesenchymal and mesenchymal to epithelial (EMT and MET) transitions. Endocrine and paracrine factors regulate EMT and MET during pregnancy. At term, increased oxidative stress forces a terminal state of EMT and inflammation, predisposing to membrane weakening and rupture. IL-6 is a constitutively expressed cytokine during gestation, but it is elevated in term and preterm births. Therefore, we tested the hypothesis that IL-6 can determine the fate of amnion membrane cells and that pathologic levels of IL-6 can cause a terminal state of EMT and inflammation, leading to adverse pregnancy outcomes.

Methods: Primary amnion epithelial cells (AECs) were treated with recombinant IL-6 (330, 1,650, 3,330, and 16,000 pg/ml) for 48 h (N = 5). IL-6-induced cell senescence (aging), cell death (apoptosis and necrosis), and cell cycle changes were studied using flow cytometry. Cellular transitions were determined by immunocytochemistry and western blot analysis, while IL-6 signaling (activation of signaling kinases) was measured by immunoassay. Inflammatory marker matrix metalloproteinase (MMP9) and granulocyte-macrophage colony-stimulating factor (GM-CSF) concentrations were measured using a Fluorokine E assay and ELISA, respectively. Amniotic membranes collected on gestational day (D) 12 and D18 from IL-6 knockout (KO) and control C57BL/6 mice (N = 3 each) were used to determine the impact of IL-6 on cell transitions. Fold changes were measured based on the mean of each group.

Results: IL-6 treatment of AECs at physiologic or pathologic doses increased JNK and p38MAPK activation; however, the activation of signals did not cause changes in AEC cell cycle, cellular senescence, apoptosis, necrosis, cellular transitions, or inflammation (MMP9 and GM-CSF) compared to control. EMT markers were higher on D18 compared to D12 regardless of IL-6 status in the mouse amniotic sac.

Conclusion: Physiologic and pathologic concentrations of IL-6 did not cause amnion cell aging, cell death, cellular transitions, or inflammation. IL-6 may function to maintain cellular homeostasis throughout gestation in fetal membrane cells. Although IL-6 is a good biomarker for adverse pregnancies, it is not an indicator of an underlying pathological mechanism in membrane cells.

Keywords: EMT; amniotic epithelial cells (AECs); cytokines; fetal membranes; inflammation.

FIGURE 1

IL-6 dose reported in term labor amniotic fluid activates JNK and p38MAPK, leading to downstream kinase activation in AECs. (A) Schematic of various phosphorylated (P) kinases tested using a multiplex immunoassay. Highlighted (yellow) P-proteins were upregulated, though not significantly, after IL-6 treatment mimicking term labor (3,300 pg/mL) compared to control-treated AECs. (B) IL-6 treatment (3,300 pg/mL) of AECs showed increasing trends of upstream kinases activation, JNK (1.7-fold) and p38MAPK (1.6-fold), which contributed to the downstream activation of MSK1 (1.5-fold), c-JUN (1.5-fold), and ATF2 (1.5-fold) compared to control samples. N = 5; mean ± SEM medium florescent intensity. (C,D) IL-6 did not increase the inflammatory profile of AECs, as seen by no changes in MMP9 activation or GM-CSF production when compared to control or term labor IL-6 treatment (3,300 pg/mL). N = 5; mean ± SEM.

FIGURE 2

IL-6 at term labor does not induce cellular aging, cell death, or cellular transitions in AECs. (A) Flow cytometry analysis of IL-6 induced senescence in AECs. Physiologic concentrations of IL-6 did not increase the population of senescence-associated-β-galactosidase (SA-β-Gal) cells compared to controls. N = 5; mean ± SEM. (B) Flow cytometry analysis of necrotic and apoptotic cells. Physiologic (330, 1,650, and 3,300 pg/mL) concentrations of IL-6 did not induce necrosis, as indicated by a lack of changes in annexin-V + propidium iodide (PI) positive cells (top right box), nor apoptosis as indicated by no changes in Annexin-V (bottom right box) expression between IL-6 treated and untreated AECs. A protein kinase panel further confirmed the lack of apoptosis by showing that phospho-p53, a pro-apoptotic marker, did not change after treatment with IL-6. Fluorescence intensity units (FIU) N = 3; mean ± SEM. (C) Immunocytochemistry of intermediate filaments in AEC to assess cell transition. AECs were stained with both cytokeratin-18 (CK-18; epithelial marker) and vimentin (mesenchymal marker) to determine transition status after physiologic (330, 1,650, and 3,300 pg/mL) IL-6 treatments. Uniform intensity analysis was conducted on vimentin and CK-18 levels to derive the vimentin/CK-18 ratio. A high vimentin/CK-18 ratio indicates a mesenchymal phenotype, while a low ratio indicates an epithelial phenotype. IL-6 treatment did not change the vimentin/CK-18 ratio compared to control cells. Fluorescent images were captured at 20×. Red: cytokeratin-18 (CK-18), green: vimentin, blue: DAPI. N = 5; mean ± SEM. Scale bar 50 μm.

FIGURE 3

IL-6 concentration reported in the amniotic fluid of infection associated pPROM activates JNK and p38MAPK leading to differential downstream kinase activation in AECs. (A) Schematic of various phosphorylated (P) kinases tested using a multiplex immunoassay. Highlighted (yellow) P-proteins were upregulated, though not significantly, after a high dose of IL-6 treatment (16,000 pg/mL) compared to control-treated AECs. (B) IL-6 treatment (16,000 pg/mL) of AECs showed increasing trends of upstream kinases activation, JNK (2.1-fold) and p38MAPK (1.9-fold), which contributed to the downstream activation of STAT1 (1.5-fold), ATF2 (1.5-fold), and HSP27 (1.5-fold) compared to control samples. N = 5; mean ± SEM medium florescent intensity. (C,D) A pathologic dose of IL-6 did not increase the inflammatory profile of AECs, as seen by no changes in MMP9 activation or GM-CSF production when compared to control or term labor IL-6 treatment (3,300 pg/mL). N = 5; mean ± SEM.

FIGURE 4

IL-6 concentration reported in the amniotic fluid of infection associated pPROM does not induce cellular aging, cell death, or cellular transitions in AECs. (A) Flow cytometry analysis of IL-6 induced senescence in AECs. Physiologic and pathologic concentrations of IL-6 did not increase the population of senescence-associated-β-galactosidase (SA-β-Gal) positive cells compared to controls. N = 5; mean ± SEM. (B) Flow cytometry analysis of necrosis and apoptosis of cells. Physiologic (3,300 pg/mL) and pathologic (16,000 pg/mL) concentrations of IL-6 did not induce necrosis, as indicated by a lack of changes in annexin-V + propidium iodide (PI) positive cells (top right box), nor apoptosis, as indicated by no changes in Annexin-V (bottom right box) expression between IL-6 treated and untreated cells. A protein kinase panel further confirmed a lack of apoptosis by showing that phospho-p53, a pro-apoptotic marker, did not change after treatment with IL-6. Fluorescence intensity units (FIU). N = 5; mean ± SEM. (C) Immunocytochemistry of intermediate filaments in AEC to assess cell transition. AECs were stained with both cytokeratin-18 (CK-18; epithelial marker) and vimentin (mesenchymal marker) to determine transition after physiologic and pathologic IL-6 treatment. Uniform intensity analysis was conducted on vimentin and CK-18 levels to derive the vimentin/CK-18 ratio. A high vimentin/CK-18 ratio indicates a mesenchymal phenotype, while a low ratio indicates an epithelial phenotype. IL-6 treatment did not change the vimentin/CK-18 ratio compared to control cells. Fluorescent images were captured at 20×. Red: cytokeratin-18 (CK-18), green: vimentin, blue: DAPI. N = 5; mean ± SEM. Scale bar 50 μm.

FIGURE 5

Normal and IL-6 KO mice undergo EMT associated changes at term. Western blot analysis and densitometry of mesenchymal marker vimentin in the amniotic sac of C57B/6 mice and KO IL-6 mice on day 12 (mid-gestation) and day 18 (term) of gestation. Amniotic sacs from control C57B/6 mice showed an increase in vimentin on day 18 compared to day 12 (11-fold), as previously document in CD1 models. The absence of IL-6 did not inhibit the increase in vimentin on day 18 compared to day 12 (7-fold). N = 3 for each category; mean ± SEM.