This research is a prospective laboratory-based study. The bs mouse (Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$) model was obtained from Jackson Laboratory (Bar Harbor, ME, USA) and raised at Shenzhen Peking University-Hong Kong University of Science and Technology Medical Center. All the animal procedures were performed in accordance with the guidelines of the ethics committee of Peking University Shenzhen Hospital. The animals were maintained under specific pathogen free (SPF) grade animal room. Approval for the study was obtained from Animal Ethics Committee of Shenzhen Peking University-Hong Kong University of Science and Technology Medical Center (Ethics approval number: 2019-205). The study with animals was carried out according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Conscious efforts were made to minimize the utilization of animal’s numbers and minimize sufferings of animals. Briefly, four-week-old mice were genotyped by sampling the tail tips. The bs model (Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$) and wild type (Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$, WT) model were prepared as previously described [17]. Female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$, Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{-}}$, and Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice were mated with proven fertile WT males, and fertility was recorded. However, in order to illustrate the effect of Tbc1d20 mutation on embryo viability, heterozygous parent mice were mated, and whether the genotype proportion of offspring deviated from Mendelian inheritance laws was calculated.

Ovaries and fallopian tubes were collected during the estrus phase for subsequent morphological analysis. Ovariectomized Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ and WT mice were prepared for investigating postnatal uterine development at two months of age. Endometrial decidualization was induced by stimulating with sesame oil. Briefly, subcutaneous injections of 17$\beta$-estradiol (E2, 100 ng/mouse; Sigma, St. Louis, MO, USA) were administered to the ovariectomized mice for 3 days (D1 to D3). After an interval of 2 days, the mice were subcutaneously injected with E2 (100 ng/mouse) and progesterone (P4) (1.0 ng/mouse; Sigma, St Louis, MO, USA) for 3 days (D6 to D8). On day 9 (D9), the left uterine horn was stimulated with sesame oil (20 µL/horn), followed by the administration of P4 for 4 days (D9 to D12). The uterine samples were collected for further analysis on D13. The collected reproductive tissues were immersed in 4% paraformaldehyde for 18 h and embedded in paraffin according to standard histological procedures. The paraffin-embedded uterine tissues were sliced into 4 µm-thick sections and mounted onto slides precoated with polylysine. The protocol was approved by the Animal Ethics Committee of Shenzhen Peking University-Hong Kong University of Science and Technology Medical Center (Ethics approval number: 2019-205).

The ovariectomized mice were injected with E2 for three days (100 ng/day) and the uterine stromal cells were isolated at D4. Briefly, uterine horn from mice was cut into 3–5 mm pieces and dissected longitudinally to expose the uterine lumen. Tissue was washed with phosphate buffered saline (PBS) and then immersed in 0.25% trypsin (Sigma, St. Louis, MO, USA) for 1 h at 4 °C followed by 1 h at room temperature. PBS containing 10% charcoal-stripped fetal bovine serum (Invitrogen, Carlsbad, CA, USA) was used to terminate digestion and mixed gently to remove epithelial clumps by pipetting. The remaining tissues were reserved, then was washed twice in PBS and finally placed in PBS containing 0.5% collagenase Type II (Sigma, St. Louis, MO, USA) for 30 min at 37 °C. 10% fetal bovine serum (FBS) was used to stop the digestion. After being washed twice in PBS, the tissues were mixed thoroughly by pipetting up and down to disperse stromal cells. The stromal cells were collected by centrifugation. The isolated primary endometrial stroma cells were first cultured in vitro for establishing a stromal cell proliferation model in 6-well cell culture plates seeded at a density of with 2.5 $\times$ 10${}^5$ cells/well. The cells were cultured in a proliferation medium comprising DMEM (Dulbecco’s Modified Eagle Medium)/Ham’s F-12 nutrient mixture (1:1) supplemented with 10% charcoal-141 stripped FBS, (Invitrogen, Carlsbad, CA, USA) and 10 U/mL penicillin-streptomycin. The in vitro decidualization cell model was generated by culturing the cells in a culture medium containing 10 nM E2 and 1 µM P4.

Sections of uterine tissue were deparaffinized and rehydrated before assigning to hematoxylin and eosin (H&E) staining or immunohistochemical assays. Immunohistochemical assays were performed using a Streptavidin-Peroxidase Histostain-Plus Kit (SAP-9100, Beijing Zhong Shan Goldenbridge Biotech., Beijing, China), according to the manufacturer’s instructions. Briefly, sections were treated with 3% hydrogen peroxide for 15 min, washed thrice with PBS, permeabilized with 0.05% Triton-X 100 for 2 min, and subsequently rinsed with PBS. Then sections were blocked with a blocking buffer for 1 h and incubated overnight at 4 °C with primary antibodies (anti-TBC1D20, 1:100 diluted, NBP1-92478, Novus Biologicals, Inc., Littleton, CO, USA; Ki67, 1:200 diluted, ab16667, Abcam, Cambridge, MA, USA; Cytokeratin, 1:200 diluted, Z0622, DAKO, Carpinteria, CA, USA; Vimentin, 1:100 diluted, ab92547, Abcam, Cambridge, MA, USA; and anti-estrogen receptor alpha antibody, 1:100 diluted, IR657, DAKO, Carpinteria, CA, USA). Section incubated with rabbit IgG was employed as negative control. Sections were subsequently incubated with the secondary antibody for 1 h at room temperature. The peroxidase reaction was performed with a SignalStain® DAB Substrate Kit (8058s, CST), and the signals were measured by light microscopy (Nikon ECLIPSE Ci-L, Nikon, Tokyo, Japan).

For immunofluorescence assay, the dewaxed uterine tissue sections were subjected to antigen retrieval in citrate antigen retrieval solution (pH 6.0), and subsequently incubated with 5% bovine serum albumin for 1 h followed by overnight incubation with the primary antibodies (anti-alpha smooth muscle actin, ab5694, Abcam, Cambridge, MA, USA; anti-CD34 antibody, PA5-89536, Thermo Fisher Scientific, Cleveland, OH, USA) at 4 °C. The tissue sections were washed with PBS and incubated with the fluorescent secondary antibody (Thermo Fisher Scientific, Cleveland, OH, USA) for 1 h at room temperature. The tissue sections were stained with 4${}^{\prime }$,6-diamidino-2-phenylindole (DAPI) for 10 min and mounted with SlowFade Gold Antifade reagent (S36936, Life Technologies, Carlsbad, CA, USA). The sections were examined with a Leica laser confocal microscope system (Leica TCS SP8, Leica, Wetzlar, Germany). For postnatal uterine morphological phenotype analysis, the results of each individual were averaged value based on three tissue sections from the proximal cervical segment, the middle segment of the uterine horn and the tail segment of the uterine horn of the same uterus, respectively.

For postnatal uterine morphological and histological measurement, the results from H&E assay, immunohistochemistry analysis and immunofluorescence assay of each individual, were average valued based on three tissue sections from the proximal cervical segment, the middle segment of the uterine horn and the tail segment of the uterine horn of the same uterus, respectively. To measure the thickness of the longitudinal and circular layers of the myometrium, starting from the side of the uterine mesentery, lines perpendicular to the circumference of the longitudinal and circular layers of the uterus at multiple evenly distributed locations was drawn and valued by image J (https://imagej.net/ij/), a public domain Java image processing program inspired by National Institutes of Health (NIH) Image for the Macintosh. The calculate average value was deemed as the thickness of thickness of the longitudinal and circular layers of the myometrium.

The total RNA was extracted and purified from tissues/cells using TRIzol™ reagent (Invitrogen, CA, USA). The concentration of the RNA was quantitated using an UV5Nano Spectrophotometer (Mettler-Toledo, LLC, Columbus, OH, USA). The RNA was reverse transcribed to the complementary DNA (cDNA) using a PrimeScript RT Master Mix kit (Takara, Dalian, Liaoning, China) with 1 µg of total RNA per reaction in a final volume of 20 µL was used. Quantitative RT-PCR (qRT-PCR) was performed using the SYBR Premix Ex Taq II PCR Kit (Takara, Dalian, Liaoning, China), according to the manufacturer’s 185 instructions with an Applied Biosystems™ 7500 Real-Time PCR System (Thermo Fisher Scientific, Cleveland, OH, USA). The protocol of qRT-PCR was as follows: pre-denaturation at 95 °C for 30 sec, denaturation at 95 °C for 5 sec, and 35 cycles of annealing at 60 °C for 34 sec, following which the reaction was halted at 4 °C. The results of PCR were analyzed using the comparative threshold cycle ($\mathrm{\Delta }$$\mathrm{\Delta }$Ct) method and the expression levels were normalized to those of Gapdh transcription level. The sequences of the primer pairs used for qRT-PCR are enlisted in Supplementary Table 1.

The proteins were extracted from whole cells using lysis buffer (50 mM Tris-HCl (pH 7.6) and 2% sodium dodecyl sulfate (SDS)) supplemented with proteinase inhibitor cocktail (P8340, Sigma-Aldrich, St. Louis, MO, USA). The concentrations of the proteins were measured using the bicinchoninic acid (BCA) method (Thermo Fisher Scientific, Waltham, MA, USA). The protein extracts (30 µg per sample) were separated by 12% SDS-PAGE (polyacrylamide gel electrophoresis) and electrophoretically transferred onto polyvinylidene difluoride (PVDF) membranes (0.45 µm; Millipore, Bedford, MA, USA). The PVDF membranes were blocked with 5% skim milk in Tris-buffered saline with 0.05% Tween-20 (TBST) for 1 h at 24 °C, and sequentially incubated overnight with the primary antibodies at 4 °C. The membranes were then incubated with a horseradish peroxidase (HRP)-conjugated secondary antibody for 1 h at 24 °C. The signals were measured using a MilliporeSigma™ Immobilon™ Western Chemiluminescent HRP Substrate (Billerica, MA, USA) and the results were recorded with an Amersham ImageQuant™ 800 Western Blot Imaging System (Cytiva, Marlborough, MA, USA). Information regarding the primary antibodies used for western blotting are provided in Supplementary Table 2.

Blood samples were collected from two-month-old Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ and Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ female mice at dioestrum, and the serum samples were obtained, aliquoted, and stored at –80 °C until further analyses. The serum levels of progesterone (P4), estrogen (E2), and follicle-stimulating hormone (FSH) were measured by RIA method. The assay was performed by Beijing North Institute of Biological Technology (Beijing, China) by running the Xh1080 RIA counter (Xi’an Nuclear Instrument Factory, Xian, Shanxi, China), and each sample comprised six different individuals per genotype. Estradiol radioimmunoassay kit (B05TB, Beijing North Institute of Biotechnology Co., Ltd., Beijing, China) for serum E2 measurement (detection range, 6–1000 pg/mL; sensitivity, 5 pg/mL), Progesterone radioimmunoassay kit (KIP1458, Beijing North Institute of Biotechnology Co., Ltd., Beijing, China) for P4 measurement (detection range, 0.12–36 ng/mL; sensitivity, 0.05 ng/mL), and Follicle-stimulating hormone immunoradiometric assay kit (B03, Beijing North Institute of Biotechnology Co., Ltd., Beijing, China) for serum FSH measurement (detection range, 1.5–100 mIU/mL; sensitivity, 0.2 mIU/mL) was employed in serum hormones test.

Cell proliferation was detected using a Cell-Light™ EdU (5-ethynyl-2-deoxyuridine) Apollo567 In Vitro Kit (RiboBio, Guangzhou, Guangdong, China) according to the Kit Operation Manual. Briefly, cell was seeded in 96-well plate. Culture medium containing EdU was added to well and incubation with cells for 1 h. Then the cells were fixed with polyformaldehyde (PFA) for 10 min at room temperature. The cell nuclei were double dyed with EdU and DAPI for assessing cell proliferation. The images were captured with a laser confocal microscope (Leica TCS SP8), and the cells were counted from six different microscopic fields.

The data were presented as the mean $\pm$ standard error of mean (SEM) and analyzed with GraphPad Prism 9.0 (FAQ 2176, GraphPad, San Diego, CA, USA). The Chi-square test was used for determining whether the number of newborn mice with different genotypes conformed to the Mendelian laws of inheritance. The statistical differences of observed value between the WT and mutation null mice were assessed using Student’s t-test, and p ˂ 0.05 was considered statistically significant.

The mutation in Tbc1d20 results in a p.Phe231Met substitution followed by an in-frame p.Arg232_Val235 deletion in the Tbc1d20 protein (Supplementary Fig. 1A), which disrupts the highly conserved amino acids in the TBC domain of Tbc1d20. Result of statistical assay of reproductive capacity showed that Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ female mice were infertile (Table 1). However, there were no significant differences in fertility between the heterozygous mutant and WT mice (Table 1).

In order to explore the causes of infertility in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ female mice, we examined the function of the ovaries and fallopian tubes at estrus, which were the first decision factor in time course of female reproductive capacity. As depicted in Fig. 1A–D, there were no obvious histological changes of ovary and fallopian tube between the Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ and WT mice. We also observed the occurrence of implantation in the uterus of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice at day 6 of gestation (Fig. 1E). However, the number of implantations was significantly reduced compared to Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice (Fig. 1F). The Chi-square test revealed that the different genotypes of newborn mice was significantly deviated from Mendel’s laws of inheritance with reduced number of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ genotype survival until birth (Supplementary Table 3). The size of the uterus of two-month-old mutant mice was remarkably smaller than that of the WT mice at diestrus (Supplementary Fig. 1B), which implicated the important role of TBC1D20 in regulating postnatal uterine development in mice.

Fig. 1.

Female mice with loss-functional mutations in Tbc1d20 exhibited normal ovulation and successful embryonic implantation. (A–D) Hematoxylin and eosin (H&E) staining of ovaries and uterine samples during the estrus phase in Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice. Bar, 100 µm. (E) Embryonic implantation in female Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice at day 6 of gestation. Bar, 5 mm. (F) Comparison of the number of implantation sites of female wild type (WT) and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice; n = 6, *p 0.01.

The uterine histomorphology of the mutant Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice was measured. Ovariectomy was performed for eliminating the interference of ovarian hormones in mice uterine at the age of 2 month. Two weeks later, the uteri were collected from the cycle-synchronized mutants and WT mice, and the uterine weights were found to be significantly lower in the mutant mice (mutant vs. WT was 0.022 g vs. 0.035 g, p = 0.0006, n = 6; Fig. 2A,B). The results of histological assays also revealed that the cross-sectional areas of the uteri of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ bs mice were smaller than those of WT mice (Fig. 2C). Moreover, the number of endometrial glands in each observed field differed significantly between the mutants and WT mice (p = 0.0033, n = 6; Fig. 2D,E). Structure of the myometrium was determined by measuring the expression of alpha-smooth muscle actin ($\alpha$-SMA), which is a biomarker of myofibroblasts. Results of the immunofluorescence assays demonstrated that the thickness of the longitudinal (p = 0.0009, n = 6) and circular layers (p = 0.0065, n = 6) of the myometrium in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice was thinner than those of the WT mice (Fig. 2F–H). The endothelial marker CD34, was employed for labeling the uterine vessels (Fig. 2I). The number of vessels between the circular layer of the myometrium and the stromal layer of the endometrium were significantly limited in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice than in the WT group (p = 0.0009, n = 6; Fig. 2J, marked by a white dashed line). A similar change was also observed in the number of vessels between the longitudinal and circular layers (p = 0.021, n = 6; Fig. 2K).

Fig. 2.

Loss-functional mutations in Tbc1d20 impair postnatal uterine development in mice. (A–C) H&E staining revealed a reduction in the size of the uterus in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice, and the cross-sectional area of the uterus and uterine weights were significantly lower than those of the WT mice (n = 6). (D,E) The number of uterine glands was reduced in female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice compared to that of the WT mice. (F–H) Representative images of immunofluorescence due to $\alpha$-smooth muscle actin ($\alpha$-SMA) (marker of myofibroblasts) depicting that the thickness of the longitudinal and circular myometrium layers, indicated by red and yellow lines, respectively, was significantly restricted in female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice. (I–K) Representative images of immunofluorescence due to CD34, a marker of vascular endothelial cells, depicting a decrease in the number of uterine vessels in female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice. The white dashed line marks the boundary between the uterine stromal cell layer and the uterine circular muscle layer; n = 6, ** p 0.01. Bar, 100 µm.

Concentration of sex hormones, the E2, P4 and FSH in maternal serum at the diestrus stage, were determined by RIA. Results from RIA indicated that the levels of E2 (p = 0.0095, n = 6; Fig. 3A) and P4 (p = 0.0133, n = 6; Fig. 3B) decreased significantly in female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice compared to those of WT mice, but there were no significant differences about the serum levels of FSH between Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ and WT mice (p = 0.2273, n = 6; Fig. 3C).

Fig. 3.

Loss of function mutations in Tbc1d20 affect the expression of steroidal sex hormones in female mice. The concentrations of steroidal sex hormones, including estrogen (E2) (A), progesterone (P4) (B), and follicle-stimulating hormone (FSH) (C) were measured by radioimmunoassay (RIA) (n = 6). The expression levels of these hormones in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mutant mice were compared to those of WT mice. **p 0.01.

Results from our research indicated that the affected ovary function and embryo implantation ability does not seem to be the main cause of infertility in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice. Thus, it was valuable to check whether the infertility of female Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice could be mainly attributed to other reasons, such as defects in uterine decidualization. In this study, we assessed the differentiation ability of the endometrial stromal cells into decidual stromal cells using an animal model of in vivo artificial decidualization (Fig. 4A). Decidual response was induced in the WT mice, but not the Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice through sesame oil injection (Fig. 4B). The ratio of the weights of the stimulated horns to the unstimulated horns of the same individual was significantly lower in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice than WT mice (Fig. 4C). Immunohistochemistry (IHC) assay exhibited that the expression of TBC1D20 increased significantly in the uterus of Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice along with decidualization (Fig. 4D). Moreover, the increasing expression of Tbc1d20 was also proved by the results of qRT-PCR assays (Fig. 4E). Ki67, an indicator of proliferation, was highly expressed in the tissues of Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice subjected to artificially induced decidualization; however, the Ki67 positive cells were very rare in the uterine tissues of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice (Fig. 4G). Analysis of the expression of cytokeratin (CK) and vimentin, revealed that the degree of decidualization in the endometrial tissues of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice was much lower than that of Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice (Fig. 4G). Notably, the expression of estrogen receptor alpha (ER$\alpha$) was limited in the uterus of decidualized Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice (Fig. 4G). The results of qRT-PCR demonstrated that the mRNA levels of the key decidualization factors, such as Bmp2, Bmp4, Hoxa10, Pgr and Wnt4, were significantly elevated in the uterus of Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ mice compared to those of the Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ group following the induction of decidualization (Fig. 4F).

Fig. 4.

Loss-functional mutations in Tbc1d20 results in failure of uterine decidualization. (A) Procedure of artificial decidualization in vivo. (B) Morphology of the uterine tissues in which the left horn was stimulated with sesame oil and the right horn was left untreated as the normal control. (C) Ratio of the average weights of the stimulated uterine horns to those of the unstimulated uterine horns from WT and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice (n = 6); **p 0.01. (D) Immunohistochemistry analysis revealed that TBC1D20 was abundantly expressed in the uterine GE, LE, and My of female mice on the first day of artificial decidualization and in the decidua on day 12. Serial sections incubated with rabbit IgG instead of the primary antibody were used as the normal control (NC). LE, luminal epithelium; GE, glandular epithelium; My, myometrium. (E) The levels of Tbc1d20 mRNA increased significantly in the oil-stimulated left horn that underwent decidualization compared to that of the unstimulated horn of WT mice. (F) The expression of the marker genes of decidualization, Bmp2, Bmp4, Hoxa10, Pgr, and Wnt4, were reduced in the stimulated uterine horns of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice compared to that of the WT mice. (G) Histological analysis of representative uterine sections from WT and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mutant mice. Representative images of Ki67, cytokeratin (CK), vimentin, and estrogen receptor alpha (ER$\alpha$) immunohistochemistry in the oil-stimulated uterine tissues of WT and Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice; n = 6. **p 0.01. Bar, 100 µm.

Stromal cells were isolated from the uterus of diestrous mice and cell proliferation was detected with 5-Ethynyl-2’-deoxyuridine (EdU) assays, for 1 h in vitro. The findings revealed a significant decrease in the number of EdU-positive uterine stromal cells lacking functional TBC1D20 (Fig. 5A,B). The expression levels of cyclin B1 and cyclin D3 decreased significantly in the proliferative uterine stroma cells of WT mice while the expression levels of cyclin D1 increased significantly (Fig. 5C,D) compared to those of the mutant mice. These findings suggested the occurrence of impaired DNA synthesis in the uterine stromal cells of Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ mice, which presented as G1/S phase arrest during cell proliferation. An in vitro decidualization model of murine uterine stromal cells was generated by the addition of E2 and P4 in the culture medium, and results demonstrated that the expression of vimentin was higher in the TBC1D20-depleted cells in the decidualization model. The findings indicated that TBC1D20 deficiency inhibited the induced decidualization of the uterine stromal cells of mice in vitro (Fig. 5E). Furthermore, the expression level of Bmp2, Bmp4, Hoxa10, Pgr and Wnt4 mRNA, were also significantly elevated in Tbc1d20${}^{\mathrm{+}\mathit{}\mathrm{/}\mathrm{+}}$ decidualization induction uterine stroma cells when compared to Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ (Fig. 5F).

Fig. 5.

Mutations in Tbc1d20 retarded the proliferation and decidualization of uterine stromal cells in vitro due to the disrupted function of endoplasmic reticulum. (A,B) The 5-Ethynyl-2’-deoxyuridine (EdU) incorporation assay revealed that Tbc1d20 depletion significantly inhibited the proliferation of endometrial stromal cells. (C) Western blotting was used to detect the levels of cyclin B1, cyclin D1, and cyclin D3 proteins in murine endometrial stromal cells during the proliferative phase in vitro. (D) The results of quantitative reverse transcription polymerase chain reaction (qRT-PCR) revealed that the deletion of Tbc1d20 inhibited the expression of cyclin B1 and cyclin D3 mRNA in murine endometrial stromal cells during the proliferative phase in vitro. (E) Western blotting was used to detect the levels of vimentin protein in murine endometrial stromal cells that underwent decidualization following artificial induction. (F) The expression level change of five marker genes associated with uterine stroma cell decidualization differentiation were detected by qRT-PCR. (G) Western blotting was used to detect the expression levels of endoplasmic reticulum stress-related proteins, namely, CHOP, BiP, IRE$\alpha$, PDI, and PERK, P-PERK, eIF2$\alpha$ and P-eIF2$\alpha$ in murine endometrial stromal cells during the proliferation and decidualization phases in vitro. **p 0.01. Bar, 100 µm.

In our previous study, we demonstrated that the deletion of TBC1D20 resulted in endoplasmic reticulum stress in mouse Sertoli cells, and excessive endoplasmic reticulum stress induced DNA damage and promoted cell death. We therefore examined the expression of the key proteins that regulate endoplasmic reticulum stress. The results demonstrated that the expression levels of C/EBP homologous protein (CHOP), inositol-requiring 1 alpha (IRE1$\alpha$), PRKR-like endoplasmic reticulum kinase (P-PERK) and phosphorylation of the $\alpha$ subunit of eukaryotic translation initiation factor 2 (P-eIF2$\alpha$) increased significantly in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ endometrial stromal cells in both the proliferative and differentiation phases. The expression levels of binding-immunoglobulin protein (BIP) and protein disulfide-isomerase (PDI) decreased significantly (Fig. 5G) in Tbc1d20${}^{\mathrm{-}\mathit{}\mathrm{/}\mathrm{-}}$ endometrial stromal cells in the proliferative and differentiation phases. The findings also indicated that the primary stromal cell from uterine tissues of mice with TBC1D20 deficiency exhibited severe endoplasmic reticulum stress. These findings indicated that the deletion of TBC1D20 induces excessive endoplasmic reticulum stress in endometrial stromal cells and inhibits cell proliferation and differentiation that leads to the failure of uterine decidualization.