Endometrial extracellular matrix rigidity and IFNτ ensure the establishment of early pregnancy through activation of YAP

Abstract

Background: In mammals, early pregnancy is a critical vulnerable period during which complications may arise, including pregnancy failure. Establishment of a maternal endometrial acceptance phenotype is a prerequisite for semiheterogeneous embryo implantation, comprising the rate-limiting step of early pregnancy.

Methods: Confocal fluorescence, immunohistochemistry and western blot for nuclear and cytoplasmic protein were used to examine the activation of yes-associated protein (YAP) in uterine tissue and primary endometrial cells. The target binding between miR16a and YAP was verified by dual-luciferase reporter gene assay. The mouse pregnancy model and pseudopregnancy model were used to investigate the role of YAP in the maternal uterus during early pregnancy in vivo.

Results: We showed that YAP translocates into the nucleus in the endometrium of cattle and mice during early pregnancy. Mechanistically, YAP acts as a mediator of ECM rigidity and cell density, which requires the actomyosin cytoskeleton and is partially dependent on the Hippo pathway. Furthermore, we found that the soluble factor IFNτ, which is a ruminant pregnancy recognition factor, also induced activation of YAP by reducing the expression of miR-16a.

Conclusions: This study revealed that activation of YAP is necessary for early pregnancy in bovines because it induced cell proliferation and established an immunosuppressive local environment that allowed conceptus implantation into the uterine epithelium.

Keywords: IFNτ; Mechanoresponses; YAP; endometrium; pregnancy.

FIGURE 1

YAP is activated in the endometrium of pregnant bovines. A, Representative bovine uterine epithelia labelled with EpCAM antibody (epithelial marker), FOXA2 antibody (glandular‐specific marker), and DAPI (nuclei) in a uterine tissue section. Higher magnification of the glandular tissue indicated in the dotted square in the lower merge panel is shown in the lower panel. Scale bars: 200 μm (up), 400 μm (medium) and 20 μm (lower). B, Representative histopathological images from uterine tissue sections in pregnant and non‐pregnant cows. Scale bars: 400 (Left) and 50 μm (Right). C, E‐cadherin visualized by immunostaining (green); nuclei were counterstained with DAPI (blue), n = 2. Scale bars, 50 μm. D, Protein expression of E‐cadherin, YAP and p‐YAP in bovine uterine tissue by western blot, n = 3. E, Western blotting for YAP in nuclear and cytoplasmic protein fractions from bovine endometrium, n = 3. F, G, mRNA levels of YAP (f), CTGF and ANKRD1(g) in bovine endometrium were detected by RT‐qPCR, n = 3. H, Representative immunohistochemical images (left, n = 2) and quantification of YAP positive cells (right, n = 3). Scale bars 200 μm. I, Representative immunofluorescence (n = 2) and quantifications of nuclear and cytoplasmic subcellular localization of YAP (right, n = 15) in bovine endometrial epithelial cells. Scale bars 100 μm. Experiments were repeated n times with two biological replicates. Data are shown as the mean ± SEM. P values were determined by an unpaired two‐sided t test. **P < .001. See also Figure S1

FIGURE 2

YAP is regulated by cell density in a manner dependent on the Hippo pathway. A, Images of primary cultured bovine endometrial cells (left) and immunofluorescence images corresponding to cell type markers (right). bEECs, bovine endometrial epithelial cells, CK18; bESCs, bovine endometrial stromal cells, vimentin. Scale bars, 100 μm and 50 μm, n = 3. B, Schematic representation of different bEEC density culture methods and images of actual effects. The same number of cells were seeded in culture plates of different sizes. C, Protein expression of YAP and p‐YAP in bEECs at different cell densities, n = 3. D, Immunofluorescence images (left, n = 2) and quantification of the nuclear and cytoplasmic subcellular localization of YAP (right, n = 15) in bEECs, when grown at low/high densities. Scale bars, 50 μm. Enlarged = 10 μm. E, YAP expression in the cytosol and/or nucleus of bEECs grown at low/high density, n = 3. F, mRNA levels of CTGF and ANKRD1 in bEECs cultured at low/high density were detected by RT‐qPCR, n = 3. G, Protein expression of YAP in bEECs treated with siNC or siLATS at low/high density, n = 3. H, Representative immunofluorescence of YAP in bEECs treated with siLATS at low/high densities, n = 2. Scale bars, 50 μm. DAPI, blue, nuclei; F‐actin, green, cell boundaries; YAP, red. All experiments were repeated independently three times with similar results, n = 3. Experiments were repeated n times with two biological replicates. Data are shown as the mean ± SEM. P values were determined by an unpaired two‐sided t test. **P < .001. See also Figure S2A–C

FIGURE 3

YAP is regulated by ECM stiffness independent of the Hippo pathway and requires tension of the actin cytoskeleton. a. Schematic representation of bEECs plated on hydrogels with different rigidities (40/1 kPa). B, Western blotting for YAP in nuclear and cytoplasmic protein fractions from bEECs plated on 40 kPa and 1 kPa fibronectin‐coated hydrogels for 48 hour, n = 3. C, Confocal immunofluorescence images (left, n = 2) and quantifications of nuclear and cytoplasmic subcellular localization (right, n = 15) of YAP in bEECs plated on hydrogels with different rigidities. Scale bars, 10 μm. D, RT‐ qPCR analysis of mRNA levels of CTGF and ANKRD1 in bEECs plated on hydrogels with different rigidities, n = 3. E, Protein expression of YAP in bEECs transfected with siNC or siLATS at 40/1 kPa hydrogels, n = 3. F, Representative immunofluorescence of YAP in bEECs transfected with si LATS and cultured on hydrogels, n = 2. Scale bars, 50 μm. G, Confocal immunofluorescence images (left, n = 2) and quantification of nuclear and cytoplasmic subcellular localization (right, n = 15) of YAP in bEECs plated at 40 kPa or low density. Cells were also treated with the F‐actin inhibitor latrunculin A (Lat.A, 0.5 μM) or PBS (control) for 24 hour. Scale bars, 20 μm. h. RT–qPCR of bEECs grown under low or high conditions on the indicated hydrogels, n = 3. Experiments were repeated n times with two biological replicates. Data are shown as the mean ± SEM. P values were determined by an unpaired two‐sided t test (c, d, g) and two‐way ANOVA (h). **P < .001. DAPI, blue, nuclei; F‐actin, green, cell boundaries; YAP, red. See also Figure S2d–h

FIGURE 4

IFNτ mediates the regulation of YAP by decreasing the expression of miR‐16a. A, Protein expression levels of YAP detected in bEECs treated with 10, 50, or 100 ng/ml IFNτ for 24 hour by western blot, n = 3. B, Immunofluorescence images (left, n = 2) and quantification of YAP positive cells (right, n = 5) in bEECs treated with 100 ng/ml IFNτ. Scale bars, 50 μm. C, Heatmap showing commonly expressed miRNAs with significant expression variance. The colour scale indicates relative expression levels of miRNAs. CT and CS: PBS treatment for 12 hour and 6 hour, respectively. TS and TT: IFNτ treatment for 12 hour and 6 hour, respectively. D, Volcano plot for abnormal expression of miRNAs after 12 hour of bovine epithelial cells treated with 100 ng/ml IFNτ. Differentially expressed miRNA were exhibited a 2‐fold change in expression with an adjusted P value of 0.05. E, RT‐qPCR analysis of relative miR‐16a expression levels in bEECs treated with 100 ng/ml IFNτ, n = 3. F, Schematic diagram showing dual‐luciferase reporter constructs harbouring the 3′‐UTR of YAP with the putative miR‐16a‐binding site. G, Luciferase activity was measured using the dual‐luciferase reporter assay system, n = 3. H, Expression of YAP protein after treatment with the indicated regimen, n = 3. I, Confocal immunofluorescence images (n = 2) and quantification of YAP positive cells (right, n = 15) in bEECs transfected with mimics‐NC or miR‐16a mimics and treated with IFNτ (100 ng/ml) or PBS. J, Expression levels of miR‐16a in bEECs transfected with siNC, siYAP and pcDNA3.1(+) YAP were detected using RT‐qPCR, n = 3. Experiments were repeated n times with two biological replicates. Data are shown as the mean ± SEM. P values were determined by an unpaired two‐sided t test (B, E), one‐way ANOVA (I, J) and two‐way ANOVA (G). *P < .05, **P < .01. See also Figure S3

FIGURE 5

YAP activation can provides the physiological environment needed for early pregnancy. A, CCK‐8 kits were used to assess proliferation of bEECs transfected with siNC, siYAP or pcDNA3.1(+) YAP at 0, 12, 24, 48 and 72 hour. *P < .05, **P < .01. B, Immunofluorescence images (left) and quantifications of Ki67positive cells (right) in bEECs transfected with siNC, siYAP or pcDNA3.1(+) YAP at 40 kPa/1 kPa hydrogels, n = 3. Scale bars, 200 μm. **P < .01, ^# P < .05, ^## P < .01. C, Immunofluorescence images of E‐cadherin and vimentin (left, n = 2) and quantification of vimentin⁺ positive cells (right, n = 5) in bEECs transfected with YAP overexpression plasmid or siYAP at 48 hour. Scale bars, 200 μm. E‐cadherin, red; DAPI, blue; Vimentin, green. *P < .05, **P < .01. D, Protein expression levels of YAP, E‐cadherin, vimentin and vinculin were detected in bEECs, n = 3. E, RT‐qPCR analysis of relative IL‐6 expression levels in bEECs transfected with siYAP or pcDNA3.1(+) YAP, n = 3. F, Representative immunofluorescence of YAP in bEECs treated with IL‐6 (10 ng/ml) or PBS at 24 hour, n = 2. G, RT‐qPCR analysis of relative LIF and VEGF expression levels in bEECs transfected with siYAP or pcDNA3.1(+) YAP, n = 3. **P < .01. Experiments were repeated n times with two biological replicates. Data are shown as the mean ± SEM. P values were determined by an unpaired two‐sided t test (e, g) and two‐way ANOVA (A–C). See also Figure S4

FIGURE 6

YAP is activated by mechanical and chemical cues in mouse pregnancy‐related models. A, Timeline of embryo development and euthanasia. B, Whole uterus images and uterine sections stained with H&E. Scale bars, 500 μm. C, Immunofluorescence detection of YAP endometrial sections from pregnant mice. Scale bars, 400 μm, n = 2. D, Protein expression levels of YAP and p‐YAP were detected in mouse uterine tissue, n = 3. E, RT‐qPCR analysis of YAP, CTGF and ANKRD1 expression levels, n = 3. F, Schematic outline of the establishment of pseudopregnancy mouse model. G, Schematic illustration of pseudopregnant mouse model. Whole uterus image and uterus sections stained with H&E. Scale bars, 500 μm. H, Effect of oil on uterine weight change 4 d. Data are shown as the fold change compared to controls without oil injection, n = 6. I, Immunofluorescence images of YAP showing uterine cross‐sections from pseudopregnant mice 4 d after intrauterine injection of oil, n = 2. Experiments were repeated n times with two biological replicates. Data are presented as the mean ± SEM. P values were determined by unpaired two‐sided t tests (h) and one‐way ANOVA (e). *P < .05, **P < .01. See also Figure S5

FIGURE 7

YAP is required for embryo implantation in vivo. A, Schematic illustration of intrauterine injection of siYAP or intraperitoneal injection verteporfin. B, Protein expression levels of YAP and p‐YAP were detected in mouse treated with VP or PBS (control), n = 3. C, Representative image of embryo implantation sites from the control and VP groups at ED 8.5. D, The number of implanted embryos observed in each uterine horn on day ED 8.5. Control, n = 14; VP, n = 9. E, F, Costaining of BrdU with YAP (e) and quantification of BrdU⁺ positive cells (f) in mouse endometrial sections, n = 2. Scale bars, 200 μm. G, RT‐ qPCR analysis of mRNA levels of IL‐6, HOXA 10 and LIF in the mouse uterus, n = 3. H, Effect of YAP knockdown on implantation rate in mice with an image showing implantation sites (arrows) in the control uterine horn (right, siNC) compared to the siYAP treated samples. I, Western blotting data showing uterine YAP levels after injection of siNC or siYAP, n = 3. J, The number of implanted embryos observed in each uterine horn on day ED 8.5. n = 6. K, Costaining of BrdU with YAP and quantification of positive cells (right) of BrdU⁺ in mouse endometrial sections, n = 2. Scale bars, 100 μm. Experiments were repeated n times with two biological replicates. Data are mean ± SEM. P values were determined by an unpaired two‐sided t test (D, E, I) and one‐way ANOVA (F). **P < .01