IL17A Suppresses IGFBP1 in Human Endometrial Stromal Cells

: Interleukin (IL) 17A has been implicated in preeclampsia, preterm labor, and miscarriage. IL17A production in non-lymphoid tissues is mainly carried out by unconventional γδ 17T cells. Innate lymphoid cells (ILCs) 3, a subgroup of innate lymphocytes, can also be a source of IL17A in the endometrium and are required from implantation to early pregnancy, with their regulation ensuring that pregnancy continues. Herein, we examined the expression of γδ 17T cells and ILC3 regulators IL1B , IL23A , and IL17D and IL17A receptors ( IL17RA / IL17RC ) in human endometrial stromal cells (EnSCs) and cell lines (KC02-44D). Accordingly, quantitative polymerase chain reaction and immunoblotting were employed. IL1B , IL23A , and IL17D were significantly upregulated in decidualized EnSCs and KC02-44D cells. A significant augmentation in IL17RA / IL17RC was also observed in decidualization. IL17A stimulation of KC02-44D cells during decidualization suppressed the decidualization marker IGFBP1 . The involvement of transcription factor Forkhead box protein O1 (FOXO1) in this repression was reflected by its translocation from the nucleus into the cytoplasm. A role for IkB kinase alpha in FOXO1 phosphorylation-mediated migration was also suggested. Taken together, our findings indicate that the secretion of IL17A by γδ 17T and ILC3 cells in the uterus contributes to EnSCs function and may play critical roles in regulating IGFBP1-mediated implantation and fetal growth.


Introduction
The human menstrual cycle takes approximately 28 days, during which the proliferation of the functional layer, secretory changes in epithelial cells, stromal cell decidualization, and menstruation occur periodically within the endometrium [1].After the follicle transforms into the corpus luteum, the uterine gland secretion is activated by progesterone, which is produced by the corpus luteum.Progesterone also acts on endometrial stromal cells (EnSCs) to induce morphological and functional differentiation, known as decidualization, thus providing a microenvironment suitable for implantation of the embryo in humans.Upon decidualization, EnSCs produce and secrete cytokines, chemokines, and growth factors, such as insulinlike growth factor binding protein 1 (IGFBP1), angiogenic factors, and prolactin (PRL) [2].Abnormal decidualization leads to placental dysplasia and underdevelopment of the spiral artery, which can cause subsequent implantation failure and miscarriage [2][3][4].
In early pregnancy, the embryonic and maternal cells have been found to interact [5].The prediction of receptor-ligand molecular pairs in the placenta and decidua of women at 6-14 weeks of gestation indicated the presence of pregnancy-specific cell subsets and revealed that EnSCs and immune cells in the decidua interact with embryo-derived extravillous trophoblasts.Data suggest that the inflammatory cytokine interleukin (IL)17A may have a pathological role in preeclampsia, premature birth, and miscarriage, with previous studies having focused on IL17A produced by immune cells at the maternal-fetal interface [6][7][8].IL17A is part of the IL17 family, which plays important roles in immune responses and host defense [9,10].Interestingly, members of this family have been proposed to have unique and non-overlapping functions against cancer, autoimmune disease, and infection.IL17A, the first member of this family to be identified, plays a critical role during infection.Persistently high levels of IL17A and homologous proteinIL17F induce inflammation.Meanwhile, IL17B has been implicated in tumorigenesis [11].IL17C influences mucosal barrier integrity and is implicated in autoimmune diseases [12,13].IL-17D is a newly identified cytokine, whose receptor was recently identified as CD93 [14,15].IL-17E (IL-25) enhances allergic inflammatory responses and is involved in host defense [16][17][18].A dysregulation of IL17A has been reported in unexplained recurrent pregnancy loss [19][20][21][22].There are emerging reports that endometrial IL17A promotes trophoblast migration [23,24].
Originally, IL17A was thought to be produced mainly by Th17 cells, with a subsequent report showing that IL17A-secreting Th17 cells induced human trophocyte invasion in first-trimester trophocytes [25].However, γδ17T cells have recently attracted attention as an alternative source of IL17A in non-lymphoid tissues, including the uterus [26][27][28].Murine uterine γδT cells produce high levels of IL17A; therefore, they are called γδ17T cells [23].Further, the reduction of Th17 cell reactivity and reversal of the Th17/Treg imbalance did not improve pregnancy outcomes in recurrent implantation failure, suggesting that Th17 cell-derived IL17A exerts minimal effects in the endometrium.Innate lymphoid cells (ILCs) 3 are another potential source of IL17A within the endometrium, as they express ROR7t [29][30][31][32].Although ILC3 almost disappear from the decidua in late pregnancy [30,33], an increase in ILC3 numbers in the decidua causes early delivery [34], suggesting the need for ILC3 production between implantation and early pregnancy as well as its suppression during continued pregnancy.Decreased ILC production or abnormal functional interactions of ILCs with the human decidua may cause fetal death [35].Despite a marked increase in ILC3 within the endometrium during the implantation window being associated with autoimmune thyroid disease and female infertility, the function of ILC3 has been reported to be suppressed in that period [36].Thus, we hypothesized that γδ17T cells and ILC3 may be regulated by endometrium-derived cells, as well as other immune cells, including uterine natural killer (uNK) cells specifically found in the endometrium [33,37].
In this study, we observed for the first time the upregulation of IL1B, IL23A (which enhances the production of IL17A by γδ17T [38][39][40] and ILC3 [41]), and IL17D (which specifically regulates ILC3 function via the CD93 receptor [14,15]) in decidualized EnSCs.Based on the increase in IL17A receptor expression during decidualization, we investigated whether IL17A secreted by activated-γδ17T and ILC3 could affect decidualization through a feedback mechanism.IL17A downregulated decidualization marker IGFBP1 in decidualized EnSCs and the KC02-44D cell line and regulatory mechanisms for IGFBP1 downregulation by Forkhead box protein O1 (FOXO1) phosphorylation.

Ethical Statement
The study was explained to all eligible patients, and we obtained informed consent from all participants.The Kansai Medical University review board approved this study (ID: 2006101), and it was conducted in accordance with the Helsinki Declaration.Human uteri with a benign myoma were donated by six patients (42-50 years) with regular menstrual cycles (Table 1).Patients who received preoperative hormone therapies were not included in this study.Histologically normal endometria were obtained from all subjects.

Human EnSC Cell Line KC02-44D and Treatment
Because patient-derived EnSCs can vary greatly in their responsiveness to stimulation, owing to the influence of many confounding factors, including differences in the timing of collection during the menstrual cycle, the expression of cytokines that may regulate IL17A-producing cells was examined using KC02-44D cells (American Type Culture Collection, Manassas, VA, USA), an established human EnSCs cell line [45].KC02-44D were cultured in DMEM medium without phenol red containing Glutamax (2 mmol/L, Thermo Fisher Scientific), 10% Fetal Bovine Serum (EU Origin), Charcoal Stripped (CS-FBS, Biowest, France), streptomycin (100 µg/mL), and penicillin (100 IU/mL).Cells were cultured until near confluence, changing the medium every 3 days.KC02-44D were treated with 10 −8 mol/L E2, 10 −7 mol/L MPA, and 0.5 mM 8-Bromo-cAMP (Sigma-Aldrich Co., LLC, St. Louis, CO, USA) (E2 + MPA + cAMP treatment) for up to 6 days for the induction of decidualization [46].Unstimulated cells were prepared as a control group.To test the effects of IL17A on decidualization, in addition to the control and E2 + MPA + cAMP treatments, 10 ng/mL IL17A (Recombinant Human IL-17A, Fujifilm Corp., Tokyo, Japan) or E2 + MPA + cAMP + 10 ng/mL IL17A stimulations were conducted for up to 6 days.

Quantitative Polymerase Chain Reaction (qPCR)
The extraction of total RNA from EnSCs or KC02-44D cells with or without treatment was conducted using the Sepasol ® -RNA I Super G (Nacalai Tesque Inc., Kyoto, Japan).Reverse transcription was performed using a ReverTraAce qPCR RT master mix with gDNA remover (Toyobo, Osaka, Japan).qPCR analysis was conducted on a LightCycler96 (Roche Diagnostics K.K., Tokyo, Japan) and Thunderbird Next qPCR Mix (Toyobo).The qPCR primers are presented in Table 2. Relative expression was calculated using the 2 −∆∆Ct method [47].Hypoxanthine-phospho-ribosyl-transferase 1 (HPRT1) was used as a housekeeping gene.

Western Blotting
To determine whether IL17A stimulation led to FOXO1 phosphorylation, we examined the downstream pathways of IL17A.IL17A binds to the IL17RA homodimer, IL17RC homodimer, and/or IL17RA-IL17RC heterodimer [48] and activates intracellular ACT1 by changing its steric structure, subsequently transmitting the signal to IkB kinase alpha (IKKA)/IKKB [49].In contrast, the downstream regulation of Akt and FOXO1 phosphorylation by IKKA has been previously described [50].

Immunocytochemistry
The sterilized cover glass was laid on a 6-well plate, and 0.3 × 10 6 KC02-44D cells were seeded onto it.Three groups were studied: (I) control, (II) E2 + MPA + cAMP-treated, and (III) E2 + MPA + cAMP + 10 ng/mL IL17A-treated groups.After 6 days with medium change every 3 days, cultured KC02-44D were subjected to formaldehyde fixation for 5 min at 22 • C and were then permeabilized with methanol for 5 min at 4 • C. KC02-44D were then incubated with a primary antibody (Foxo1-rabbit-mAb (C29H4) #2880 diluted to 1/1000, Cell Signaling Tech, RRID: AB_2106495) overnight at 4 • C, followed by incubation with a goat anti-rabbit-Alexa488 (# A-11008, Molecular Probes, US-OR, RRID: AB_143165) diluted 1/3000 and 1 µg/mL propidium iodide.LSM700 (Carl Zeiss Co., Ltd., Tokyo, Japan) was used to observe the stained cells.Positive staining rates in the nucleus and cytoplasm were calculated and averaged by visually counting the cytoplasm-stained, nucleus-localized, and cytoplasm-localized FOXO1 in images obtained from two independent individuals.In total, 15 images (945 cells) were taken in group I, 15 images (552 cells) in group II, and 15 images (552 cells) in group III of an area covering 1 mm 2 .These images were used for analysis.

Statistical Analyses
The normality of data was confirmed using the Shapiro-Wilk test.Welch's t-test was used to compare the means of the two groups with the Bonferroni correction.p-values of < 0.05 were considered as indicative of statistical significance.SPSS software was used for all statistical analyses (version 22.0; IBM Corp., Armonk, NY, USA).

Elevations in IL1B, IL23A, and IL17D in Decidualized EnSCs
After 12 days of decidualization treatment of primary culture EnSCs, IL1B and IL23A were significantly upregulated compared with their levels in unstimulated cells (control) (p < 0.05) (Figure 1).Further examination of changes in the expression of IL17 gene family members revealed no expression of IL17A, IL17E (IL25), or IL17F in EnSCs.In contrast, IL17B, IL17C, and IL17D were detected, and a significant upregulation of IL17D was observed during decidualization (p < 0.05, Figure 1).

IL17A Treatment of KC02-44D Cells
As IL17RA and IL17RC expression was detected in EnSCs, IL17A secreted from activated γδ17T and ILC3 cells might have some effect on decidualization.Therefore, we treated KC02-44D cells for 6 days with 0, 0.1, 1.0, 10, or 100 ng/mL IL17A, based on previous studies [37] to examine changes in PRL.A significant increase in PRL was observed in the decidualized group relative to the control group.However, no effect of IL17A was observed.Because there was no clear difference in the effect of IL17A concentration on PRL, we examined its effect on other genes using concentrations from previous studies (10 ng/mL) [52].Subsequently, KC02-44D cells were subjected to a 6-day decidualization treatment to examine changes in interleukin receptor expression.qPCR analysis confirmed a significant upregulation of IL17RA and IL17RC in decidualized KC02-44D (p < 0.05).IL22RA1 expression was not detected (Figure 3).

IL17A Treatment of KC02-44D Cells
As IL17RA and IL17RC expression was detected in EnSCs, IL17A secreted from activated γδ17T and ILC3 cells might have some effect on decidualization.Therefore, we treated KC02-44D cells for 6 days with 0, 0.1, 1.0, 10, or 100 ng/mL IL17A, based on previous studies [37] to examine changes in PRL.A significant increase in PRL was observed in the decidualized group relative to the control group.However, no effect of IL17A was observed.Because there was no clear difference in the effect of IL17A concentration on PRL, we examined its effect on other genes using concentrations from previous studies (10 ng/mL) [52].

Effect of IL17A on FOXO1 Localization in KC02-44D Cells
In KC02-44D cells, IL17A stimulation causes the phosphorylation of decidualization marker FOXO1 and its subsequent migration out of the nucleus, resulting in a downregulation of IGFBP1 [54,55].After KC02-44D cells were stimulated with or without IL17A for 6 days, we examined the subcellular localization of FOXO1 [56].We expected to stain the nucleus with propidium iodide, but the cytoplasm and nucleolus were also fluorescently labeled (Figure 5, purple).This localization suggested that the RNA molecules were labeled with propidium iodide.FOXO1 was fluorescently labeled with a FOXO1 antibody and Alexa488-secondary antibody (green).Increased localization of FOXO1 in the nucleus was found in the decidualized cells in comparison with that in the untreated cells, with changes in FOXO1 localization to the cytoplasm observed following additional IL17A

Effect of IL17A on FOXO1 Localization in KC02-44D Cells
In KC02-44D cells, IL17A stimulation causes the phosphorylation of decidualization marker FOXO1 and its subsequent migration out of the nucleus, resulting in a downregulation of IGFBP1 [54,55].After KC02-44D cells were stimulated with or without IL17A for 6 days, we examined the subcellular localization of FOXO1 [56].We expected to stain the nucleus with propidium iodide, but the cytoplasm and nucleolus were also fluorescently labeled (Figure 5, purple).This localization suggested that the RNA molecules were labeled with propidium iodide.FOXO1 was fluorescently labeled with a FOXO1 antibody and Alexa488-secondary antibody (green).Increased localization of FOXO1 in the nucleus was found in the decidualized cells in comparison with that in the untreated cells, with changes in FOXO1 localization to the cytoplasm observed following additional IL17A stimulation (Figure 5A).FOXO1 localization in the nucleus or cytoplasm was determined, and the frequency of FOXO1 relative to the total cell number included in the entire image was calculated for each image, whereafter differences in frequency were assessed.A significant increase in the nuclear localization of FOXO1 was found in decidualized cells and IL17A additionally stimulated cells in comparison with that in untreated KC02-44D controls (p < 0.05).Furthermore, a significant increase in FOXO1 translocation from the nucleus to the cytoplasm was observed in the IL17A additionally stimulated cells in comparison to that in decidualized cells (p < 0.05) (Figure 5B).The cytoplasm and nucleolus were stained with PI (purple).FOXO1 was fluorescently labeled using a FOXO1 antibody and an Alexa488-secondary antibody (green).The high-power images are magnified versions of the dashed squares in the lowpower images (A).Localization in the nucleus or cytoplasm was determined, whereafter the frequency was calculated relative to the number of cells in the entire image covering an area of 1 mm 2 (B).* p < 0.05 vs. control using Welch's t-test with Bonferroni correction.+ p < 0.05, vs. decidualization using Welch's t-test with Bonferroni correction.IL17A, interleukin 17 A; FOXO1, Forkhead box protein O1; PI, propidium iodide.

Discussion
In this study, decidualization of EnSCs isolated from the human endometrium and of the endometrial stromal cell line KC02-44D induced a significant upregulation of IL1B, IL23A, and IL17D expression [15,41], affecting the proliferation, differentiation, and functional regulation of γδ17T and ILC3 [30,33,[38][39][40]57,58], which is involved in innate immunity and embryo implantation in the endometrium.This finding prompted us to explore the interactions between decidualized EnSCs and IL17A-producing immune cells.Therefore, we examined the expression of IL17RA/IL17RC and IL22RA1, which encode receptors for IL17A and IL22 [32] secreted from activated γδ17T and ILC3, observing the upregulation of IL17RA and IL17RC, but not IL22RA1, in both EnSCs and KC02-44D cells.Furthermore, IL17A stimulation during decidualization of KC02-44D cells suppressed IGFBP1 expression.As the involvement of FOXO1 in this repression has been suggested [59], the intracellular localization of FOXO1 was examined.FOXO1 migration into the cytoplasm was observed in decidualized cells with additional IL17A stimulation.IGFBP1 downregulation occurred following FOXO1 migration out of the nucleus.Furthermore, analysis of IL17A receptor downstream signaling revealed a reduction in the amount of IKKA, suggesting the involvement of IKKA in this localization change.
It has been suggested that γδT cells are the major source of IL17A in various nonlymphoid tissues, including the uterus [23,[26][27][28].The crucial role of IL17A in maintaining health in response to infection, physiological stress, and injury, as well as in autoimmune diseases, is well recognized [38,[60][61][62].γδT cells, which are thought to be a key source of IL17A in the uterus [23], may facilitate IL17A-induced trophoblast migration and thus play an important role in defense against infection, in addition to potential involvement in pregnancy in young women.Unlike γδT cells in other tissues, uterine immune cells need to both defend against infection and tolerate allogeneic fetuses during pregnancy.γδ17T cells in the endometrium do not express granzyme B and CD107a like cytotoxic γδT [23].Thus, γδ17T cells are thought to tolerate the fetus, with cytotoxic γδT providing protection against infection.Comparison of data from allogenic and syngeneic reproductive pairs also indicates that γδT cells may play an important role in successful pregnancy [23].Furthermore, γδT cells have been found to be abundant at the maternal-fetal interface during pregnancy as well as in the non-pregnant uterus [23], suggesting their involvement in endometrial decidualization.In the decidualized EnSCs in this study, IL17A was implicated in the regulation of IGFBP1 expression, a major decidualization marker.Overexpression of IGFBP1 in the endometrium has been associated with impaired fetal growth and placental insufficiency and is thought to cause placenta accreta [63].Our results suggest that in decidualized EnSCs, IL17A decreases IKKA protein levels, regulates downstream Akt and FOXO1 phosphorylation, and restores IGFBP1 expression to a constant level.This suggests that IL17A secreted by γδ17T and ILC3 may contribute to the prevention of fetal growth defects caused by placental insufficiency and the development of placenta accreta by restricting the overexpression of IGFBP1 in decidualized EnSCs.
ILCs are immune cells involved in innate defense and tissue remodeling.NK cells, lymphoid tissue inducer cells, and ILC1-3 belong to the ILC family [64], with all subsets observed within human uterine tissues [29][30][31]33].The ILC3 fraction identified via flow cytometry also contains ILC precursors that can give rise to ILC1-3 [65,66].ILC3 in the human decidua interact with decidualized EnSCs, suggesting that they contribute to innate immunity as well as vascular and tissue architecture [33].Indeed, ILC3 in the decidua likely act on extravillous trophoblasts and modulate invasion by secreting granulocytemacrophage colony-stimulating factor, X-C motif chemokine ligand 1, and macrophage inflammation protein 1α and 1β [31,67,68].The results of this study suggest that their secretion from ILC3 [68] is dependent on activation by IL1B, IL23A, and IL17D secreted from decidualized EnSCs.
IL17A, from intrauterine γδ17T cells, which is thought to play an important role in EVT invasion, is highly abundant [23].γδ17T cells are recruited into the uterus in response to estrogen, which increases in the blood with follicle development and induces IL17A production in γδ17T cells [69].Therefore, their involvement in uterine function from the proliferative to secretory phases is also possible.In this context, it is interesting to follow the changes in expression and secretion of IL1B, IL23, and IL17D in EnSCs under estrogen stimulation conditions.ILC3 was reduced in eutopic endometrium during endometriosis [70] when compared to levels in normal endometrium.Therefore, we suggest the involvement of ILC3 in normal decidualization and the menstrual cycle.Further, dysregulation of ILC3 might contribute to endometriosis, consequent implantation failure, and recurrent pregnancy loss.

Limitations
This study demonstrated that the levels of IL1B, IL23A, and IL17D, which regulate γδ17T and ILC3 proliferation and differentiation, are elevated in decidualized EnSCs.Further, IL17A secreted from these cells affects EnSCs decidualization.However, because γδ17T and ILC3 samples derived from the human endometrium are difficult to obtain and these cell lines have not been established outside of mice [71,72], examining the effect of supernatant from decidualized human EnSC culture, which is thought to contain IL17D and other factors, on the growth and differentiation of human cells has not been possible.Since IL17D is also expressed in CD19(+) B cells and quiescent CD4(+) T cells [73,74], determining the localization of IL17D expression in EnSCs using actual uterine tissue sections is necessary.Furthermore, a comparison of IL1B, IL23A, and IL17D expression in decidualized EnSCs and the presence of γδ17T or ILC3 in the vicinity of decidualized EnSCs using endometrial tissue sections from healthy subjects and patients with infertility may lead to the discovery of a new category of obstetric diseases caused by γδ17T and/or ILC3.It is also necessary to measure cytokines, including IL17A and others, in the blood of patients experiencing preeclampsia, preterm labor, and miscarriage.

Conclusions
We found that EnSCs may regulate γδ17T and ILC3 in the endometrium and that IGFBP1 is suppressed by IL17A secreted by the latter two cell types.This regulation occurs via the IL17A-IKKA-FOXO1 phosphorylation-IGFBP1 axis.

Figure 5 .
Figure5.Effect of IL17A on FOXO1 in KC02-44D.The cytoplasm and nucleolus were stained with PI (purple).FOXO1 was fluorescently labeled using a FOXO1 antibody and an Alexa488-secondary antibody (green).The high-power images are magnified versions of the dashed squares in the lowpower images (A).Localization in the nucleus or cytoplasm was determined, whereafter the frequency was calculated relative to the number of cells in the entire image covering an area of 1 mm 2 (B).* p < 0.05 vs. control using Welch's t-test with Bonferroni correction.+ p < 0.05, vs. decidualization using Welch's t-test with Bonferroni correction.IL17A, interleukin 17 A; FOXO1, Forkhead box protein O1; PI, propidium iodide.

Figure 5 .
Figure5.Effect of IL17A on FOXO1 in KC02-44D.The cytoplasm and nucleolus were stained with PI (purple).FOXO1 was fluorescently labeled using a FOXO1 antibody and an Alexa488-secondary antibody (green).The high-power images are magnified versions of the dashed squares in the low-power images (A).Localization in the nucleus or cytoplasm was determined, whereafter the frequency was calculated relative to the number of cells in the entire image covering an area of 1 mm 2 (B).* p < 0.05 vs. control using Welch's t-test with Bonferroni correction.+ p < 0.05, vs. decidualization using Welch's t-test with Bonferroni correction.IL17A, interleukin 17 A; FOXO1, Forkhead box protein O1; PI, propidium iodide.

Figure 6 .
Figure 6.Mechanism of IL17A-dependent FOXO1 migration.After KC02-44D cells were stimulated with or without 10 ng/mL IL17A for 6 days, IKKA, IKKB, and ACTB levels were quantified (A).The obtained signals were analyzed using ImageJ and normalized to ACTB band intensity.Significant suppression of IKKA was noted with IL17A stimulation ((B), p < 0.05).There was no significant difference in IKKB among the three groups (C).* p < 0.05 vs. control using Welch's t-test with Bonferroni correction.IL17A, interleukin 17 A; IKKA (CHUK), component of inhibitor of nuclear factor