The involvement of the progesterone receptor in PIBF and Gal‐1 expression in the mouse endometrium

The progesterone‐regulated genes, PIBF and Gal‐1, are key players in the feto‐maternal immunological interaction. This study aims to investigate the expression of PIBF and Gal‐1 in WT and progesterone receptor KO models as well as subsequent effects of PIBF on decidualization of stromal cells.


| INTRODUC TI ON
The biological activities of progesterone (P) are mainly mediated by two nuclear progesterone receptor (PR) isoforms: PRA and PRB. [1][2][3] Mice lacking PRA are infertile, 4,5 while engagement of PRB accounts for mammary gland development. The infertility of mice missing both isoforms (PRKO mice) is partly due to the inability of endometrial cells to undergo decidual transformation and to reach the receptive state. 4 The relative expression of PRA and PRB proteins is likely to be critical to ensure appropriate reproductive tissue responses to P. 5 Studies on PR knockout mice revealed that PRs are required not only for endometrial receptivity and decidualization but also for establishing an appropriate immune environment in the endometrium. 4,6 In vivo studies on the role of oestrogen (E) and P in regulating the uterine immune environment showed that the proinflammatory effect of E is antagonized by P in the uteri of wildtype (WT) mice, but not in those of progesterone receptor knockout (PRKO) mice. 6 Further, PRs in the thymus are required for thymic involution during pregnancy and for normal fertility, 7 showing that PRs are involved in modulating the function of the immune system on several levels.
Progesterone regulates the genes of theprogesterone-induced blocking factor (PIBF) and of galectin-1 (Gal-1), both playing a role in the feto-maternal immunological interaction. 8 Gal-1-deficient mice show an increased rate of fetal loss. P treatment increases Gal-1 expression in myometria and deciduae of stress-challenged pregnancies at gestational day 7.5. Gal-1 treatment on the other hand prevents the stress-induced decrease of both progesterone and PIBF levels and normalizes the resorption rates. 9 These data suggest the existence of a progesterone-mediated mechanism that regulates Gal-1 expression at the feto-maternal interface.
A nucleus-associated 90 kDa PIBF isoform is constitutively expressed in immature and rapidly proliferating cells, for example, embryonic tissues and tumours. [10][11][12][13] Smaller PIBF isoforms are expressed by various pregnancy-associated tissues. Both villous and extravillous trophoblast cells express 30, 50 and 90 kDa PIBF isoforms in first trimester. 14 The latter are localized in the cytoplasm, and after being secreted, decrease NK activity and induce Th2 cytokine production. Anti-PIBF treatment of pregnant mice results in pregnancy loss. If, however, NK cells are depleted by the use of anti-NK antibodies, PIBF deficiency does not cause an increased rate of resorptions. 15 Decidual NK also appear to be affected by PIBF, which inhibits upregulation of perforin expression in activated decidual lymphocytes as well as NK cell cytotoxicity by blocking degranulation. 16,17 In addition to causing pregnancy loss, the lack of PIBF alters the cytokine balance. Neutralization of endogenous PIBF activity in pregnant mice by specific anti-PIBF antibody reduces the synthesis of IL-10 and increases that of IFN-γ. 18 The PIBF receptor is a glycosylphosphatidylinositol (GPI)-anchored protein, which, for signalling, temporarily associates with the alpha chain of the IL-4 receptor, 19 thus PIBF signals via the Jak/STAT pathway.
Both P and PIBF play a role in the induction of the Th2-biased cytokine balance. IL-4 or PIBF treatment of peripheral pregnancy lymphocytes results in immediate STAT6 phosphorylation, whereas a 24-hour continuous presence of progesterone is required for the same effect, suggesting that progesterone might activate the Jak/ STAT pathway indirectly, via first inducing PIBF. 19 Along that line, PIBF1 gene was identified as P-regulated gene by cistrome analysis of the murine uterus using chromatin immunoprecipitation (ChIP) followed by massive parallel sequencing (ChIP-seq). 20 Taken together, these data suggest that by upregulating Th2type cytokine production and by down-regulating NK activity PIBF affects the immune response in a way, which might have an impact on the feto-maternal relationship.
Though later stages of pregnancy have been relatively well characterized in this respect, little is known about the involvement of PIBF in the embryo-maternal interactions during early pregnancy.
This study was aimed at investigating the expression of PIBF and Gal-1 in wild-type and PR KO mice, and the subsequent effect of PIBF on stroma cell decidualization. were ovariectomized and allowed to rest for 2 weeks. Mice were then treated with 1 mg of P subcutaneously, or sesame oil as a control, and killed 6 hours later. Uterine tissues were collected for RNA isolation and immunohistology.

| Mice
For investigating pregnancy-induced mRNA and proteins expression, wild-type BALB/c female mice (10-12 weeks old) were housed with fertile males. The day, the vaginal plug was observed, was considered day 0.5 of pregnancy. The animals were killed by cervical dislocation on specific days of pregnancy (dpc), and uterine tissue was used for mRNA isolation and immunohistochemical analyses.

| Primary stromal cell culture
Isolation of primary stromal cells was performed as described previously. 21,22 In order to standardize the phase of the oestrous cycle, 6-to 8-week-old mice were subcutaneously injected for three consecutive days with 100 ng/100 µL of E, prior to stromal cell isolation. On the 4th day, uterine horns of E treated mice were dissected longitudinally Four hundred thousand cells were seeded in 6-well cell culture plates containing collagen-coated coverslips (collagen I from rat tail, 50 μg/mL in 0.02 mol/L acetic acid, BD biosciences) and incubated at 37°C and 5% CO 2 . The unattached cells were removed by washing several times with HBSS, and fresh medium (control) or media supplemented with 2 µg/mL of PIBF was added. The culture medium was changed every 2 or 3 days with continuous supplementation with PIBF.
The level of desmin expression was used as a marker for decidualization. 23

| RNA isolation and real-time RT-PCR analysis
Total RNA was isolated from uterine tissues (n = 3 per pool) using  All Images were captured on Olympus imaging system BX51 equipped with DP71CCD camera (Olympus, Tokyo, Japan), and CellF imaging software was used. Images were edited using Photoshop CS6 (Adobe, San Jose, CA, USA).

| Statistical analysis
All data are reported as mean ± SEM and were analysed by a Student's t-test and one-way ANOVA followed by the Scheffe post hoc method. P ≤ 0.05 was considered significant. All statistical analyses were performed using STATISTICA 10 (StatSoft Inc, Tulsa, OK, USA).

| In the mouse uterus, PIBF is produced via activation of progesterone receptor A
To test whether activation of the classical PRs was involved in PIBF production, 6-to 8-week old ovariectomized BALB/c mice were treated with 1 µ g of P, or sesame oil as a control, for 6 hours. Uterine tissues were collected for RNA isolation and immunohistology. Real-time reverse-transcription PCR was performed for quantification of gene expression.
PIBF mRNA was significantly increased in ovariectomized and progesterone (P)-treated wild-type (WT) mice, but not in PRKO and PRAKO mice under the same conditions ( Figure 1A). PIBF expression in the uterus was checked by immunohistochemistry. PIBF protein expression was reduced in the endometria of PRKO and PRAKO, but not in PRBKO mice ( Figure 1B).
These data suggest that PIBF transcription depends on the activation of PR-A.

| PIBF induces decidual transformation of mouse endometrial stromal cells
To investigate the possible role of PIBF in decidualization, endometrial stromal cells were isolated from the uterus of non-pregnant mice.
The cells were cultured to confluence and incubated with medium F I G U R E 1 PIBF expression in the uterus of ovariectomized, P-treated mice. Six-to eight-wk-old BALB/c mice were ovariectomized and allowed to rest for 2 wk. Mice were then treated subcutaneously with 1 mg of P, or sesame oil as a control, and killed 6 h later, when uterine tissues were collected for RNA isolation and immunohistology. (A) Real-time reverse-transcription PCR was performed for quantification of gene expression and normalized to expression level in PRKO mice. The Pibf1 mRNA was increased in ovariectomized and progesterone-treated WT mice, but not in PRKO and PRAKO animals. The bars represent the mean ± SEM of three experiments. *P ≤ 0.05 (Student's t-test). (B) Immunohistochemical analysis of P-induced expression of PIBF in the endometria of WT, PRAKO and PRBKO mice. Five-micrometre paraffin sections were reacted with 1:25 diluted biotinylated monoclonal anti-PIBF antibody. PIBF protein expression was reduced in the endometria of PRKO and PRAKO, but not in PRBKO mice. Images were captured on an Olympus imaging system equipped with a DP71CCD camera with 40× objective. This experiment was performed three times with identical results F I G U R E 2 PIBF induces decidualization of mouse endometrial stromal cells. Endometrial stromal cells were isolated from the uterus of non-pregnant mice. Decidualization was induced by incubating confluent cells for 6 d in a (A) medium without PIBF and (B) with PIBF (2 μg/ mL). The level of desmin expression was used as a marker for the decidualization. The cells were incubated overnight at 4°C with 1:200 diluted rabbit anti-desmin antibody, and Alexa 1:300 diluted donkey anti-rabbit 488 nm was used to visualize immune complexes. Images were captured on an Olympus imaging system equipped with a DP71CCD camera with 100× objective. This experiment was performed three times with identical results. Desmin reactivity appears as green fluorescence. Nuclei were counterstained with DAPI (blue)

(A) (B)
containing 2 μg/mL PIBF for 6 days. Control cell cultures were incubated in medium without PIBF. Desmin was used as a marker for decidualization. 23 During a 6-day culture, PIBF induced significantly elevated desmin expression in mouse decidual stromal cells (Figure 2), which indicates that PIBF might be one of the factors that contribute to decidua formation, and possibly endometrial receptivity.

| PIBF expression in the uterus corresponds with the implantation window
To assess the expression profile during the peri-implantation period,

| The expression of Gal-1 mRNA and Gal-1 in the uteri of BALB/c mice during early pregnancy
Previous findings suggest that Gal-1 expression is required for implantation and the maintenance of early pregnancy. We tested Gal-1 expression at the implantation sites in early pregnancy.
Pregnant BALB/c mice were killed at the indicated time points, uterine horn tissue was removed, and the conceptuses on day 7.5 and 11.5 of pregnancy were carefully dissected out of the implantation site. Real-time reverse-transcription PCR was performed for quantification of gene expression ( Figure 4A). Gal-1 expression in the endometria was detected by immunofluorescence, using 1:400 diluted rabbit polyclonal antibody to Gal-1 (Abcam) followed by donkey anti-rabbit Alexa Fluor 594 nm (1:500 dilution; Molecular Probes) as a second antibody ( Figure 4B).
Both Gal-1 mRNA and Gal-1 expression continuously increased in the mouse uterus between day 2.5 and day 11.5 of gestation.

| ERα, PR and Gal-1 expression in decidual NK cells
Earlier data show that murine decidual NK cells express PIBF in their cytoplasmic granules, 26 and that there is a cross-regulation between progesterone and Gal-1 at the feto-maternal interface. 9 Therefore, we tested whether decidual NK cells express PRs and Gal-1.
Decidual NK cells do not express PRs ( Figure 5A), but most of them do express ERα ( Figure 5B) and Gal-1 at day 7.5 murine pregnancy at the mesometrial area of the endometrium ( Figure 6).
These findings confirm that PRs are absent from decidual NK cells. Therefore, it is unlikely that either PIBF or Gal-1, localized in the cytoplasmic granules of these cells, is produced by the NK cells themselves, rather taken up from the surrounding environment.

| D ISCUSS I ON
The Pibf1 gene contains a progesterone response element 20 and is activated following engagement of PRA in the mouse uterus.
Earlier data revealed that anti-PIBF treatment of pregnant mice at day 10 results in an increased resorption rate, indicating that PIBF is required for the maintenance of pregnancy. 27 F I G U R E 3 PIBF expression in the endometrium of wild-type mice during the peri-implantation period. BALB/c mice were mated overnight. Sighting of the vaginal plug was considered day 0.5 of pregnancy. Animals were killed at the indicated time points. Uterine horns were removed, incubated with 4% PFA, paraffin-embedded sections (5 µm) were prepared, and the slides were reacted with 1:25 diluted anti-PIBF monoclonal antibody. Images were captured on an Olympus imaging system equipped with a DP71CCD camera with 10× (left panels) and 40× objective (right panel). This experiment was performed three times with identical results Recently, we showed that extracellular vesicles produced by the pre-implantation mouse embryo express PIBF, and that PIBF containing extracellular vesicles can alter the cytokine profile of murine spleen cells. 28 These data suggest that PIBF might play a role in the feto-maternal interaction during the peri-implantation period.
PIBF induces decidualization of mouse endometrial stromal cells, and most importantly, its expression in the mouse uterus corresponds with the implantation window.It is well documented that P activity is required for the establishment and maintenance of pregnancy, 29,30 and the present observations suggest that not only some of the pro-gestational effects of P are mediated by PIBF but also that PIBF might indeed play a role in the process of implantation.
We demonstrate a continuous increase in the mRNA expression of another P-regulated gene Gal-1 between days 2.5 and 11.5 of gestation in mice. Gal-1 gene expression in the mouse uterine tissues is regulated by ovarian steroids during implantation 8 ; furthermore, a cross-regulation between Gal-1 and P has been demonstrated in stress-challenged murine pregnancies. 9 Although Gal-1 expression in the female reproductive system was first reported in the 1990s, many functional aspects of this lectin during pregnancy have recently been discovered. [31][32][33] Gal-1 is present on trophoblast cells in the placental bed in haemochorial placentation and decidualization. 34,35 Proteomic studies showed that Gal-1 expression is reduced on placental villous tissues from patients with spontaneous miscarriages. 36 In mice, Gal-1 deficiency results in a higher frequency of fetal loss in allogeneic matings compared to syngeneic matings. 37 In our hands, a subpopulation of decidual NK cells expressed Gal-1. Though all NK cell types express Gal-1, 38 Gal-1 + NK cells are enriched in the decidua, compared to peripheral blood, 38 and Gal-1 is endowed with an immune-regulatory function. 39 In spite of their high perforin content, spontaneous cytotoxic activity of decidual NK cells is moderate. 40 Though there is no evidence that NK cells directly attack the trophoblast, recurrent miscarriage is often accompanied by increased decidual NK activity, suggesting that this mechanism might be a component of the underlying pathology. [41][42][43][44][45] In mice, PIBF protects pregnancy by controlling NK activity. 24 Earlier data 16 show that PIBF inhibits degranulation of NK cells.
Decidual NK cells contain PIBF, which co-localizes with perforin in the cytoplasmic granules. In day 12.5 normal mouse pregnancy, only 54% of the PIBF+ decidual NK cells are perforin positive, whereas in F I G U R E 4 Gal-1 mRNA and Gal-1 expression in uteri of wild-type mice during early pregnancy. BALB/c mice were mated overnight. Sighting of the vaginal plug was considered day 0.5 of pregnancy. Animals were killed at the indicated time points. Uterine horn tissue was removed and was collected for RNA isolation or immunofluorescent staining. (A) Real-time reverse-transcription PCR was performed for quantification of gene expression and normalized to expression level at 2.5 dpc. Bars represent mean ± SEM (N = 6) *P ≤ 0.05 (ANOVA followed by the Scheffe post hoc method). (B) Uterine sections were reacted with 1:400 diluted rabbit polyclonal antibody to Gal-1 followed by 1:500 diluted donkey anti-rabbit Alexa Fluor 594 nm as a second antibody. Images were captured on an . Immunocomplexes were detected with anti-rabbit Alexa Fluor 594 (red). Fluorescein-labelled DBA lectin was used to identify uNK cells (green). Nuclei were counterstained with DAPI (blue). Images were captured on an Olympus imaging system equipped with a DP71CCD camera with 40× objective. The asterix in panel B depicts the region shown at a higher magnification in the insert. This experiment was performed three times with identical results F I G U R E 6 Gal-1 expression in mouse decidual NK cells. BALB/c mice were mated overnight. Sighting of the vaginal plug was considered day 0.5 of pregnancy. Animals were killed at 7.5 dpc. For analyses of Gal-1 expression in decidual NK cells, immunofluorescent staining was performed with 1:400 diluted rabbit polyclonal antibody to Gal-1, followed by 1:500 diluted donkey anti-rabbit Alexa fluor 594 (red). Fluorescein-labelled DBA lectin (diluted 1:200) was used to identify uNK cells (green). Nuclei were counterstained with DAPI (blue). Images were captured on an Olympus imaging system equipped with a DP71CCD camera with 10× (A), 40× (B) and 100× (C) objectives, respectively. The asterix in panel B depicts the region shown in panel C at a higher magnification. This experiment was performed three times with identical results

(A) (B) (C)
mRNA splicing, angiogenesis and regulation of immune response.
Interaction of these proteins in decidual NK cells could be of importance for downstream events in pregnancy progression.

| CON CLUS ION
Our results show that activation of PRA results in PIBF production.
PIBF induces decidual transformation of stromal cells, and its expression in the mouse endometrium peeks during the implantation window, suggesting that and that PIBF might be involved in the implantation process.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.