Clusterin from endometrial glands plays a critical role in decidualization via Trem2.

CD1 mice (purchased from Guangdong Sijia Jingda Biotechnology Co., Ltd.) were housed in the SPF animal care facility under a light/dark cycle (12:12 h) and controlled temperature (22–24 °C). All animal procedures were approved by the Animal Care and Use Committee of South China Agricultural University (No. 2021f108). To establish pregnancy, female mice aged 8–10 weeks were mated with fertile mice. The pseudopregnant mice were produced by mating between the female and vasectomized males, and the day when the vaginal plugs were checked was recognized as Day 1 of pregnancy. To induce artificial decidualization, sesame oil (10 µL, Sigma‒Aldrich, St. Louis, MO, USA) was injected into one uterine horn on Day 4 of pseudopregnancy, and the noninfused contralateral horn served as the control. On Day 4, the mice were anesthetized via intraperitoneal injection of ketamine (90 mg/kg) and xylazine (10 mg/kg), followed by exposure to CO 2 . After confirming that the animals were fully anesthetized, they were euthanized by cervical dislocation, and the uteri were collected for further analysis. Ovariectomies were conducted on adult mice to examine the impacts of steroid hormones, followed by a two-week period of rest to eradicate any circulating ovarian steroids. The mice were treated with a single injection of 17β-estradiol (100 ng per mouse, Sigma‒Aldrich, St. Louis, MO, USA). The control mice were injected with vehicle. The uterus was harvested 3 h after injection. All the experiments were repeated at least three times. Frozen Sect. (10 μm) were fixed with 4% paraformaldehyde (PFA) dissolved in phosphate-buffered saline (PBS) for 10 min, followed by incubation with 0.1% Triton X-100 in PBS for 10 min. After the sections were blocked with FBS albumin in PBST for 30 min at 37 °C, they were incubated with primary antibody at 4 °C overnight. The primary antibodies used were as follows: rabbit anti-Clusterin antibody (1:200, Abcam, ab184100), rat anti-E-cadherin antibody (1:500, Abcam, ab11512), rabbit anti-FOXA2 antibody (1:400, Cell Signaling Technology, 8186); and rabbit anti-ERα antibody (1:200, Cell Signaling Technology, 13258). The next day, the cells were washed 3 times with PBST, incubated with anti-rabbit and anti-goat IgG (conjugated to Alexa Fluor 594 and Alexa Fluor 488 secondary antibodies, respectively (Jackson ImmunoResearch, William Seagrove, PA, USA)) for 30 min at room temperature, and then washed 3 times with PBST. Nuclei were counterstained with 6-diamino-2-phenylindole (DAPI, Sigma‒Aldrich, St. Louis, MO, USA). The cells were observed under a fluorescence microscope (Leica). In situ hybridization was performed as previously described [ 22 ]. Briefly, total RNA was isolated from mouse ovaries and amplified with Clu and Trem2 primers after reverse transcription. The pGEMT plasmid (Promega) was used to clone the amplified fragments of Clu and Trem2 , while primers for T7 and SP6 were used to amplify the Clu and Trem2 fragments in the pGEM- Clu / Trem2 plasmid, which were subsequently validated by sequencing. Digoxigenin-labeled antisense or sense complementary RNA probes were transcribed in vitro via a digoxigenin RNA labeling kit according to the manufacturer’s instructions. The sequences used for in situ hybridization are listed in Table  1 . Frozen 4% paraformaldehyde-fixed sections from the control and experimental groups were processed onto the same slide for 1 h and hybridized at 55 °C overnight. The sections were washed and incubated iwith alkaline phosphatase-conjugated anti-digoxigenin antibody (1:5,000). The positive signal was visualized using a buffer containing 5-bromo-4-chloro-3-indolyl phosphate (BCIP, 0.4 mM) and nitro blue tetrazolium (NBT, 0.4 mM) as reaction substrates. Endogenous alkaline phosphatase activity was inhibited with 2 mM levamisole, the samples were counterstained with 1% methyl green, and the signal was dark brown. The sequences of the RNA-targeting probes are shown in Table  1 . Table 1 Sequences of the primers used in this study Gene Name Primer Sequences Application Accession Number APOER2 GGACCTACTGACCAAGAAC GGATGAGGCGTGAATAGTT RT‒qPCR NM_001018054.3 VLDLR GACCACAGCAGTATCAGAG ATTCCGCCACATCAAGTAG RT‒qPCR NM_001018056.3 LDLR CAGCGAAGATGCGAAGAT AGAAGAGGTAGGCGATGG RT‒qPCR NM_000527.5 Clu CGAAGATGCTCAACACCTCA TCCTGCGGTATTCCTGTAGC RT‒qPCR NM_013492.3 CLU TCTTGTCTGTGGACTGTTC AGGAGGTGTTGAGCATCT RT‒qPCR NM_001831.4 TREM2 CGGCTGCTCATCTTACTCTT CAAGTTGTGCGTGCTGAC RT‒qPCR NM_001272078 RPL7 CTGCTGTGCCAGAAACCCTT TCTTGCCATCCTCGCCAT RT‒qPCR NM_011291.5 Rpl7 GCAGATGTACCGCACTGAGATTC RT‒qPCR NM_011291.5 ACCTTTGGGCTTACTCCATTGATA PRL AAGCTGTAGAGATTGAGGAGCAAA RT‒qPCR NM_000948 TCAGGATGAACCTGGCTGACTA IGFBP1 CCAAACTGCAACAAGAATG GTAGACGCACCAGCAGAG RT‒qPCR NM_001013029 Clu CGAAGATGCTCAACACCTCA TCCTGCGGTATTCCTGTAGC ISH NM_013492.3 Trem2 TGACGCCTTGAAGCACTG CCTCGGAGACTCTGACACT ISH NM_001272078 TREM2 CCCACAACACCACAGTGTT siRNA NM_001272078 Sequences of the primers used in this study GGACCTACTGACCAAGAAC GGATGAGGCGTGAATAGTT GACCACAGCAGTATCAGAG ATTCCGCCACATCAAGTAG CAGCGAAGATGCGAAGAT AGAAGAGGTAGGCGATGG CGAAGATGCTCAACACCTCA TCCTGCGGTATTCCTGTAGC TCTTGTCTGTGGACTGTTC AGGAGGTGTTGAGCATCT CGGCTGCTCATCTTACTCTT CAAGTTGTGCGTGCTGAC CTGCTGTGCCAGAAACCCTT TCTTGCCATCCTCGCCAT CCAAACTGCAACAAGAATG GTAGACGCACCAGCAGAG CGAAGATGCTCAACACCTCA TCCTGCGGTATTCCTGTAGC TGACGCCTTGAAGCACTG CCTCGGAGACTCTGACACT The stromal cell line 4003 was maintained in phenol red-free DMEM/F12 medium (Sigma) supplemented with 10% charcoal-stripped fetal bovine serum (cFBS), 500 ng/mL puromycin, 100 U/mL penicillin, 100 U/mL streptomycin, and 1% ITS (Sigma) in a 37 °C incubator containing a 5% CO 2 humidified atmosphere. To induce decidualization, the cells were treated with 1 µM medroxyprogesterone (MPA, Sigma) and 0.5 mM db-cAMP (Sigma). Recombinant CLU protein (RD, 2937) treatment was applied to decidual stromal cells. Stromal cells were seeded into 6-well plates. At 60-70% confluence, in each experimental group, the corresponding medium was replaced with fresh media containing 2% cFBS, and decidualization was induced in vitro by adding MPA (1 μm) or db-cAMP (0.5 mM), followed by treatment with or without recombinant CLU protein (1 µg per sample). The control group cells were also cultured in fresh media supplemented with 2% cFBS without MPA or db-cAMP. All of these cells were harvested after 2 days. An siRNA kit for TREM2 and nonspecific siRNA was synthesized by RiboBio Co., Ltd. (Guangzhou, China). Using Lipofectamine 3000 (Invitrogen, Waltham, MA, USA) according to the manufacturer’s protocol, a random RNA sequence not specific for any specific gene (scramble) was used as a negative control (NC). The siRNAs with the most remarkable interference efficiency (50 nM) were used for transfection into stromal cells for 24 h, after which the cells were induced to undergo in vitro decidualization. The relative mRNA levels of TREM2 were detected. The sequences of the targeted siRNAs are shown in Table  1 . RT‒qPCR was performed as previously described [ 23 ]. Mouse uterine endometrium or cultured stromal cells were harvested with TRIzol reagent (Accurate Biotechnology, Hunan, China), and RNA was extracted according to the manufacturer’s instructions. Reverse transcription into cDNA was performedwith the HiScript II Reverse Transcriptase kit (Vazyme). ChamQTM Universal SYBR ® qPCR Master Mix (Vazyme) was used to perform RT‒qPCR. The sequences of the primers used for RT‒qPCR in this study were shown in Table  1 . All experiments were repeated at least three times independently. All the data are presented as the means ± standard deviations (SDs). Statistical analyses were performed with Student’s t tests. Analysis of variance (ANOVA) was performed for multiple comparisons. p  < 0.05 was considered to indicate statistical significance. All analyses were performed with GraphPad Prism ® software (GraphPad Software Inc., San Diego, CA, USA).

The spatial expression profile of the CLU protein at implantation sites in the mouse uteri from Days 1 to 8 of pregnancy was determined by immunofluorescence. The presence of the endometrial epithelium at the implantation site was confirmed by measuring E-cadherin (E-cad) expression. CLU was strongly expressed in the endometrial glandular epithelium on Days 1 to 2 and Days 5 to 8 of pregnancy, but almost no signal was detected in the uteri on Days 3 and 4 (Fig.  1 a). As previously reported, 3D visualization confirmed that glands connect directly to the chamber during implantation and that glands become more developed and continue to elongate and stretch following implantation on Day 4.5 [ 4 ]. CLU was detected in the expanding ductal regions of glands connecting the crypt encasing the embryo on Days 5 to 8 of pregnancy (Fig.  1 a). Fig. 1 Spatiotemporal expression of Clu protein and mRNA in the mouse uteri during early pregnancy. ( a ) Immunofluorescence image showing the expression of the CLU protein in mouse uteri from Days 1 to 8 of early pregnancy. ( b ) In situ hybridization showing the expression of Clu in mouse uteri from Days 1 to 8 of early pregnancy. Scale bar: 200 μm. M, mesometrial pole; AM, antimesometrial pole. All images are representative of three independent experiments Spatiotemporal expression of Clu protein and mRNA in the mouse uteri during early pregnancy. ( a ) Immunofluorescence image showing the expression of the CLU protein in mouse uteri from Days 1 to 8 of early pregnancy. ( b ) In situ hybridization showing the expression of Clu in mouse uteri from Days 1 to 8 of early pregnancy. Scale bar: 200 μm. M, mesometrial pole; AM, antimesometrial pole. All images are representative of three independent experiments In situ hybridization revealed localized Clu mRNA expression in the endometrial glandular epithelium, similar to the results of the spatiotemporal expression pattern of immunofluorescence. In situ hybridization further confirmed the expression of Clu at implantation sites in the mouse uteri in early pregnancy (Fig.  1 b). To further corroborate whether Clu was localized to the glandular epithelium, Foxa2 was used to characterize the glands further, as it has been reported that Foxa2 is expressed only in the glandular epithelium of the uterus [ 24 , 25 ]. Immunofluorescence was used to analyze the expression of the FOXA2 and CLU. Double immunofluorescence staining revealed complete colocalization of FOXA2 and CLU proteins in the glandular epithelium from Days 5 to 8 of pregnancy (Fig.  2 a), which provided more concrete evidence of CLU expression in the glandular epithelium of the mouse uteri. Fig. 2 Clu was expressed in the glandular epithelium of the whole uterus. ( a ) FOXA2 and CLU coimmunostaining during Days 5 to 8 of early pregnancy. ( b ) CLU is expressed in the glandular epithelium at the interimplantation site of the longitudinal section of early pregnancy. ( c ) Two CLU secretion pathways exist in the mouse uteri during pregnancy. Scale bar: 200 μm. The enlarged area depicts the two CLU secretion modes: secretion to the ductal regions and secretion outside the stromal cells. Scale bar: 20 μm. Arrowheads indicate the location of the CLU. M, mesometrial pole; AM, antimesometrial pole; S, stromal cells; GE, glandular epithelium; LE, luminal epithelium. All images are representative of three independent experiments Clu was expressed in the glandular epithelium of the whole uterus. ( a ) FOXA2 and CLU coimmunostaining during Days 5 to 8 of early pregnancy. ( b ) CLU is expressed in the glandular epithelium at the interimplantation site of the longitudinal section of early pregnancy. ( c ) Two CLU secretion pathways exist in the mouse uteri during pregnancy. Scale bar: 200 μm. The enlarged area depicts the two CLU secretion modes: secretion to the ductal regions and secretion outside the stromal cells. Scale bar: 20 μm. Arrowheads indicate the location of the CLU. M, mesometrial pole; AM, antimesometrial pole; S, stromal cells; GE, glandular epithelium; LE, luminal epithelium. All images are representative of three independent experiments As a previous study showed, the glands at the interimplantation sites remain clustered together due to dilation of the implant chamber and pressure on the decidua from Days 5 to 8 of gestation [ 4 ]. Owing to the previous finding that CLU is located in the expanding glands, we hypothesized that CLU would also be detected in the glandular epithelium at interimplantation sites. To test this hypothesis, we used longitudinal sections of the implantation site and interimplantation site of the joint together and found that the glands clustered at the interimplantation site were positive for CLU expression (Fig.  2 b). These results confirmed that Clu was expressed in the glands of the whole uterus from Days 5 to 8 and not only in the glands of the implantation sites. Previous studies on Clu have demonstrated that it has two main forms, one extracellularly secreted from cells and another intracellularly retained, but the canonical pathway involves the synthesis of secretory CLU, which plays a major role in a broad range of physiological and pathophysiological functions [ 10 ]. Under enlarged visualization, we found that secretory CLU exists in two pathways and is secreted at implantation sites on Days 5 to 8 of gravidity. The results showed that in some glands, CLU was secreted into the ductal region, whereas in other glands, CLU was secreted into neighboring stromal cells through the paracrine pathway (Fig.  2 c). These data indicate that Clu has two secretion patterns in the uterus during pregnancy, which may confer correspondingly diverse functions. In both mice and rats, the estrogen peak is present on Day 4 of gestation, which is essential for the progesterone (P 4 )-primed uterus to turn into a receptive uterus [ 26 ]. If ovariectomized mice are treated with P 4 on the morning of Day 4 before the estrogen level increases, diapause of the embryo is induced, and implantation is delayed, while an injection of estrogen can induce implantation again [ 1 ]. As shown in Fig.  1 a, CLU signals were almost undetectable in the uterus on Days 3 and 4; however, they reoccurred on Day 5, so we hypothesized that estrogen might regulate the expression of CLU. The above analysis demonstrated the function of estrogen through the ER and revealed that ERα plays a more essential role in female reproduction than does ERβ [ 2 ], emphasizing the gland’s response to estrogen [ 27 ]. Thus, to test this hypothesis, we first investigated whether the glands expressed ERα, which responds to estrogen. Immunohistochemistry analysis revealed ERα in the endometrial glandular epithelium (Fig.  3 a). It was suggested that estrogen could affect glands and colocalize with CLU. Fig. 3 Clu expression is regulated by estrogen through ERα. ( a ) The colocalization of ERα and CLU on the glandular epithelium in the mouse uteri during early pregnancy was detected by immunofluorescence. Scale bar: 100 μm. ( b ) Immunofluorescence showed that CLU was induced at 3 h after treatment with estrogen in ovariectomized mouse uteri. Scale bar: 100 μm. All images are representative of three independent experiments Clu expression is regulated by estrogen through ERα. ( a ) The colocalization of ERα and CLU on the glandular epithelium in the mouse uteri during early pregnancy was detected by immunofluorescence. Scale bar: 100 μm. ( b ) Immunofluorescence showed that CLU was induced at 3 h after treatment with estrogen in ovariectomized mouse uteri. Scale bar: 100 μm. All images are representative of three independent experiments We then used an ovariectomized mouse model to further verify whether estrogen might regulate the expression of CLU. Our results revealed that the protein levels of CLU were increased by estrogen in ovariectomized mouse uteri at 3 h after estrogen administration (Fig.  3 b). Together, these studies revealed that estrogen could activate the expression of Clu in the mouse uteri in an ERα-dependent manner. The above studies have shown that members of the LDL receptor gene family, such as Lrp2 , Lrp1 , ApoER2 , and Vldlr , are responsible for the binding of Clu and that Clu acts via them [ 10 , 28 ]. Clu was identified as a crucial ligand of Trem2 from an unbiased protein microarray screen. Additionally, Clu can bind to Trem2 to clear β-amyloid in the brain. Trem2 deficiency in microglia reduces Clu uptake in mice [ 21 ]. Paracrine crosstalk between the glandular epithelium and stroma has been reported [ 3 , 29 ]. Previously, we showed that Clu can be secreted by neighboring stromal cells through the paracrine pathway. Endometrial stromal cells initiate decidualization on Day 5, with maximal decidualization occurring on Day 8. Gland stretching occurs throughout the implantation site, and Clu is expressed throughout the gland; thus, the gland may secrete Clu to neighboring stromal cells through paracrine signaling to regulate decidualization. Overall, we predicted that Clu receptors might interact with the Clu secreted from the glands to influence decidualization in endometrial stromal cells. To explore this prediction, RT‒qPCR was performed to quantify the mRNA expression levels of the LDL receptor gene family and TREM2 in the cultured endometrial stromal cell line 4003. Under in vitro decidualization, the VLDLR , LDLR , and APOER2 mRNA levels all decreased significantly, and only the TREM2 mRNA level markedly increased (Fig.  4 a). Therefore, we focused on Trem2 in the following study. To investigate whether Trem2 was expressed in vivo, we performed in situ hybridization to determine the localization of Trem2 mRNA in mouse uteri from Days 5 to 8 of pregnancy. The Trem2 mRNA signal was present in the stroma surrounding the implanting blastocyst on Day 5, at which point the zone began to form a decidua called the primary decidual zone (PDZ). The proliferating and differentiating stromal cell layer around the PDZ region subsequently expressed Trem2 until it peaked on Day 8 (Fig.  4 b). These results confirmed that endometrial stromal cells expressed Trem2 during decidualization. Fig. 4 Trem2 is expressed on stromal cells during decidualization. ( a ) RT‒qPCR analysis of Clu receptor genes in endometrial stromal cells under decidualization conditions for 2 days. The ΔΔCt method was used to calculate changes in gene expression relative to RPL7 . ( b ) In situ hybridization showing the expression of Trem2 in mouse uteri from Days 5 to 8 of pregnancy. Scale bar: 200 μm. The data are from three independent experiments. Con, control; dec, in vitro decidualization. * p  < 0.05 Trem2 is expressed on stromal cells during decidualization. ( a ) RT‒qPCR analysis of Clu receptor genes in endometrial stromal cells under decidualization conditions for 2 days. The ΔΔCt method was used to calculate changes in gene expression relative to RPL7 . ( b ) In situ hybridization showing the expression of Trem2 in mouse uteri from Days 5 to 8 of pregnancy. Scale bar: 200 μm. The data are from three independent experiments. Con, control; dec, in vitro decidualization. * p  < 0.05 To further investigate the prediction that the secretion of Clu from glands may affect decidualization through interactions with receptors, we carried out the following experiment. Intrauterine injection of sesame oil on Day 4 of pseudopregnancy in vivo can induce artificial decidualization [ 30 , 31 ]. On Day 4, the pseudopregnant mice were injected with sesame oil into the uterine lumen to induce a decidual reaction. The uterus was harvested 4 days later. Compared with the Day 4 pseudopregnant uterus, in situ hybridization revealed that Clu and Trem2 mRNA expression was distinctly increased in the glands and decidual cells, respectively, in artificial decidualization (Fig.  5 b and c). Immunofluorescence also revealed that the CLU protein was highly expressed in the glands of the artificial decidua (Fig.  5 a). Real-time RT‒qPCR was then performed to quantify the mRNA expression levels of Clu and Trem2 , which were markedly upregulated in the uterus after artificial induction for decidualization (Fig.  5 d). To verify the role of secreted Clu during decidualization, stromal cells were treated with disulfide-linked heterodimers from the recombinant CLU protein. We found that the expression levels of the widely used decidualization markers PRL and IGFBP1 in stromal cells under in vitro decidualization conditions were markedly increased by recombinant CLU (Fig.  6 c and d). To assess the importance of Trem2 during decidualization, we knocked down TREM2 expression with siRNA in cultured endometrial stromal cells in vitro. Under in vitro decidualization, compared with that under normal conditions, the TREM2 expression level was significantly lower and almost undetectable by TREM2 siRNA, confirming that the knockdown efficiency was sufficient. Then, in our analysis of the expression levels of PRL and IGFBP1 , we observed that PRL and IGFBP1 expression was obviously attenuated by TREM2 siRNA (Fig.  6 a and b). These observations indicate that the suppression of Trem2 perturbs the process of decidualization and potentially regulates Clu secretion during decidualization through Trem2 . Fig. 5 Clu and Trem2 are upregulated upon decidualization. ( a and b ) Expression of Clu in oil-induced decidualization in vivo. Immunofluorescence was used to measure the level of CLU protein in the uterus. Scale bar: 100 μm. The level of Clu mRNA was measured by in situ hybridization. Scale bar: 50 μm. ( c and d ) Expression of Trem2 in artificial decidualization in vivo. In situ hybridization revealed that Trem2 mRNA is expressed in decidual cells. Scale bar: 50 μm. RT‒PCR was performed to quantify the mRNA expression level of Trem2 . The ΔΔCt method was used to determine relative changes in gene expression with that of Rpl7 . The data are from three independent experiments. PD4, Day 4 of pseudopregnancy; AD, artificial decidualization. * p  < 0.05 Clu and Trem2 are upregulated upon decidualization. ( a and b ) Expression of Clu in oil-induced decidualization in vivo. Immunofluorescence was used to measure the level of CLU protein in the uterus. Scale bar: 100 μm. The level of Clu mRNA was measured by in situ hybridization. Scale bar: 50 μm. ( c and d ) Expression of Trem2 in artificial decidualization in vivo. In situ hybridization revealed that Trem2 mRNA is expressed in decidual cells. Scale bar: 50 μm. RT‒PCR was performed to quantify the mRNA expression level of Trem2 . The ΔΔCt method was used to determine relative changes in gene expression with that of Rpl7 . The data are from three independent experiments. PD4, Day 4 of pseudopregnancy; AD, artificial decidualization. * p  < 0.05 Fig. 6 Clu regulates decidualization via Trem2 . ( a ) Real-time RT‒PCR data showing TREM2 expression in stromal cells after the induction of decidualization in vitro. ( b ) Knockdown of TREM2 decreases PRL and IGFBP1 expression in decidual stromal cells. ( c and d ) The expression of PRL and IGFBP1 was upregulated upon recombinant CLU protein treatment in decidual stromal cells. ( e and f ) The effect of recombinant CLU protein in decidual stromal cells was blocked by TRME2 siRNA. The data are from three independent experiments. The ΔΔCt method was used to calculate changes in gene expression relative to RPL7 . Con, control; dec, in vitro decidualization. * p  < 0.05 Clu regulates decidualization via Trem2 . ( a ) Real-time RT‒PCR data showing TREM2 expression in stromal cells after the induction of decidualization in vitro. ( b ) Knockdown of TREM2 decreases PRL and IGFBP1 expression in decidual stromal cells. ( c and d ) The expression of PRL and IGFBP1 was upregulated upon recombinant CLU protein treatment in decidual stromal cells. ( e and f ) The effect of recombinant CLU protein in decidual stromal cells was blocked by TRME2 siRNA. The data are from three independent experiments. The ΔΔCt method was used to calculate changes in gene expression relative to RPL7 . Con, control; dec, in vitro decidualization. * p  < 0.05 To further examine the molecular mechanisms underlying whether the secretion of Clu impacts decidualization via Trem2 , we treated endometrial stromal cells with recombinant CLU to knock down TREM2 in vitro. Under in vitro decidualization, the expression of PRL and IGFBP1 did not increase with the silencing of TREM2 expression, which was contrary to the results obtained without TREM2 siRNA treatment (Fig.  6 c-g). Collectively, these results suggest that the secretion of Clu might mediate decidualization via Trem2 .

Successful pregnancy comprises discrete events, including implantation, decidualization, placentation and parturition. If any step of the process is disrupted, adverse ripple effects of the disruption during pregnancy result in poor outcomes [ 1 ]. Decidualization is crucial for successful gestation, necessary for blastocyst implantation, and the formation of a functional placenta [ 2 , 3 ]. Initially, attachment occurs between the blastocyst and crypt luminal epithelium, followed by proliferation and decidualization of stromal cells at the implantation site [ 4 ]. During decidualization, stromal fibroblasts are transformed into secretory decidual cells [ 5 ], which enable the endometrium to defend against excessive trophoblast invasion, act as an embryo quality sensor, and provide nutrition to the developing embryo before the placenta is fully functional [ 2 , 3 ]. Dysregulation of decidualization can result in preeclampsia or miscarriage. In rodents and primates, the uterine glands are essential for stromal decidualization. The notion that uterine glands secrete paracrine-acting factors into the stroma to induce decidualization is a novel concept derived from mouse models that are not equipped with uterine glands [ 3 ]. No evidence of implantation or stromal cell decidualization was found in progesterone-induced uterine gland knockout mice on Day 6 after mating [ 6 ]. The absence of endometrial glands resulted in a lack of several pregnancy-related factors, including leukemia inhibitory factor ( Lif ), steroid hormone receptors, cytokines, growth factors, and several developmental factors, which are critical for gestation. The infertility phenotype of Lif null mice and conditional forehead box A2 ( Foxa2 ) deletion models, as well as mice lacking uterine glands, affects gland-derived products, underscores the primary role of gland-derived products in establishing and maintaining pregnancy [ 7 ]. Clusterin ( Clu ), also known as apolipoprotein J, is a highly conserved 80-kDa disulfide-linked heterodimer glycoprotein found in various body fluids [ 8 – 10 ]. This multifunctional protein exists in two distinct forms: one is processed and matured in the endoplasmic reticulum and Golgi apparatus before being secreted into the extracellularly, and the other is secreted directly into the cytoplasm during cell stress [ 11 ]. Clu functions as a chaperone molecule and is a functional homolog of small heat shock proteins [ 12 ], binding to the hydrophobic domains of misfolded proteins and promoting their receptor family-mediated endocytosis and lysosomal degradation [ 12 , 13 ]. Since its discovery in 1983, Clu has been implicated in various physiological processes, such as cell proliferation, invasion, apoptosis, oxidative stress, stimulation of inflammatory factors, and regulation of complement activity [ 14 ]. Numerous studies conducted in the last few decades have explored the connection between Clu and Alzheimer’s disease. Clu facilitates the removal of amyloid Aβ from the blood‒brain barrier, and Clu deficiency in Alzheimer’s mice is associated with increased amyloid angiopathy in brain tissue. Additionally, previous studies have shown that Clu regulates both normal and abnormal pregnancy events. Clu has been observed to respond to hormone regulation in the uterine gland epithelium and lumen epithelium during both the human menstrual cycle and the mouse estrous cycle and is involved in physiological remodeling of the uterus. Additionally, Clu may have clearance functions mediated by receptors that promote the apoptosis of cells and the endocytosis and degradation of cell debris by lysosomes [ 15 ]. In 1996, Thomas L. Brown published a study on Clu in early pregnancy in mice; however, owing to technical limitations, the results of this study were not entirely convincing [ 16 ]. Consequently, the function and mechanism of Clu in early pregnancy remain largely unknown. Abnormal expression of Clu is associated with pregnancy-induced hypertension (PIH). Evidence suggests that the concentration of Clu in the serum increases in proportion to the severity of PIH. Clu may function as a trigger for the expression of inflammatory factors, and its dysregulation could be a crucial factor in the development of PIH [ 9 ]. Furthermore, research has indicated that irregular Clu expression is also linked to other pregnancy-related diseases, such as intrauterine growth restriction, recurrent abortion, premature delivery and preeclampsia, and may be able to predict adverse pregnancy outcomes. However, the role of Clu in regulating early pregnancy in mice under physiological conditions is not fully understood. Research has demonstrated that several members of the low-density lipoprotein receptor family (LDL receptor family), including low-density lipoprotein receptor-related protein 1 ( Lrp1) , low-density lipoprotein receptor-related protein 2 ( Lrp2 ), apolipoprotein E receptor 2 ( ApoER2 ), and very-low-density lipoprotein receptor ( Vldlr ), serve as receptors for Clu , allowing the secreted Clu to perform its physiological function [ 17 ]. For example, research as early as 1995 revealed that LRP2 binds to CLU in vitro, and cells expressing LRP2 facilitate CLU endocytosis leading to lysosomal degradation [ 18 , 19 ]. Clu amplifies central leptin signals through Lrp2 -mediated endocytosis [ 20 ]. Furthermore, ApoER2 / Vldlr -mediated Clu is implicated in inducing male germ cell meiosis [ 17 ]. Recent studies have also identified triggering receptor expressed on myeloid cells 2 ( Trem2 ) as a receptor for Clu , promoting amyloid β uptake and potentially reducing Alzheimer’s disease risk [ 21 ]. However, the receptors mediating Clu in the uterus in normal early pregnant mice remain unidentified, necessitating further research to investigate the expression, function, and regulatory mechanisms of Clu in this context.

In our study, Clu expression was observed in the uterine glands during early pregnancy. Additionally, Clu has two secretory forms in glands. Notably, estrogen regulates Clu expression in the early stages of pregnancy through the ERα pathway. Our findings also suggest a potential role for Trem2 as a mediator of Clu in the regulation of the decidual area. In conclusion, uterine gland-derived Clu , which serves as a novel paracrine modulator, may play a role in early pregnancy by influencing the decidualization process mediated by Trem2 in mice. These new findings have significant implications for understanding the mechanisms underlying successful implantation and decidualization.

Decidualization is crucial for successful pregnancy, with defects potentially leading to preeclampsia or miscarriage [ 1 ]. Foxa2 , a transcription factor uniquely expressed in uterine glands, regulates postnatal uterine gland differentiation in mice and is considered a distinct marker of these glands. Conditional knockout of Foxa2 significantly reduces the number of uterine glands, causing severe defects in endometrial decidualization [ 3 , 7 , 32 ]. Our study revealed that Clu colocalized with Foxa2 , specifically in uterine glands from days 5 to 8 of pregnancy in mice, coinciding with the timing of decidual formation. Silencing the Clu receptor Trem2 expression downregulated the expression of the decidualization markers IGFBP1 and PRL in the in vitro model of stromal cell decidualization. Treatment with recombinant CLU partially reversed this effect, suggesting that glandular-secreted Clu is involved in regulating the decidualization process. The concept of uterine glands secreting paracrine agents to promote decidualization is relatively new. In this regard, the Kazal type 3 ( Spink3 ), a gene specifically expressed in uterine gland, was revealed to be secreted into the lumen and decidual region during early pregnancy in mice, acting as a paracrine regulator to affect stromal decidualization [ 29 ]. Our experimental results showed that Clu is specifically expressed in glands and also observed in adjacent stromal cells, suggesting that Clu likely regulates decidualization through paracrine secretion. Clu regulation of decidualization may depend on two key pathways: proteostasis (clearing misfolded proteins and maintaining protein stability) and anti-inflammatory action [ 33 ]. Decidualization is a dynamic process in which endometrial fibroblasts transform into decidual cells, leading to extensive remodeling of the endometrial extracellular matrix (ECM) occurs within the uterine stroma. This ECM remodeling is essential for decidualization. Gene expression analysis and immunohistochemistry reveal that decidualization defects are associated with abnormal expression and deposition of ECM molecules [ 34 ]. Matrix metalloproteinases (MMPs) are the primary enzymes responsible for degrading collagen and other proteins during ECM remodeling. Significant changes in the expression of MMPs family molecules are observed in uterine decidual tissues [ 35 ]. The overexpression of MMP9 is observed in the uteri of women with adenomyosis and preeclampsia, and CLU, which has a strong affinity for MMP9, inhibits its enzymatic activity [ 36 ]. In inflammatory diseases, uncontrolled MMP9 activity increases, whereas CLU prevents stress-induced MMP9 aggregation and inhibits its enzymatic activity. which is associated with the inhibition of ECM deposition. Additionally, CLU inhibits the enzymatic activities of MMP2, MMP3, and MMP7. Clu knockout mice exhibit elevated ECM protein levels, underscoring the critical role of Clu in maintaining ECM balance [ 37 ]. These findings suggest that uterine gland epithelial cell-expressed Clu may be secreted into the adjacent decidualization region, and regulates decidualization by maintaining protein homeostasis and balancing the hydrolysis of ECM-related proteins in the decidualization area. Inflammatory responses are essential for decidualization in both rodents and humans [ 38 ]. This process commences with a robust proinflammatory stress response lasting several days, followed by the emergence of anti-inflammatory and senescent decidual cells [ 39 ]. Compared to undifferentiated control cells, decidualization induction for 2 days resulted in upregulation of a large number of cytokines, interleukins, and their receptors, accounting for 70 out of 84 inflammatory mediators detected by PCR array. Persistent inflammation may underlie spontaneous decidualization dysfunction [ 40 ]. Clu is also an important anti-inflammatory factor. Studies have shown that Clu can bind to its receptor proteins, thereby mitigating severe endothelial inflammatory responses [ 33 ]. Clu knockout mice display increased severity of autoimmune myocarditis, accelerated progression from acute inflammation to myocardial scarring, and elevated levels of inflammatory factors in the lungs [ 41 , 42 ]. Trem2 activation antagonizes myeloid responsiveness to proinflammatory stimuli at the signaling level [ 43 ]. The interaction between Trem2 and its ligands mitigates inflammatory responses [ 44 , 45 ]. Knockdown of Trem2 with short hairpin RNA suppresses inflammation in microglial cell cultures. These findings emphasize that Trme2 plays a role in restricting inflammation. Our data indicated a specific expression of the Clu receptor protein Trem2 in the decidualization regions of pregnant mice from days 5 to 8. The in vitro results demonstrated that Clu regulated the decidualization process through Trem2 . The potential mechanism likely involves the role of Clu and Trem2 in prevention of excessive inflammatory responses in the decidualization region. However, further investigation is required to elucidate the precise molecular mechanisms. The evidence suggests that the Clu gene responds to steroid hormone regulation in various tissues. In castrated rats, metabolic ablation of hormones results in a significant upregulation of Clu mRNA [ 46 ]. Additionally, the endometrial cancer cell line ECC-1 exhibits increased Clu expression following estrogen treatment [ 47 ]. ERα is essential for the normal development of uterine glands postnatally and for increasing the number of glands after puberty. Protein-protein interaction between ERα and FOXA2 have been confirmed by immunoprecipitation experiments, indicating that glands and their secretions are likely regulated by estrogen through ERα [ 4 ]. While the localizations of CLU and ERα proteins were overlapped on Days 5–8 of early pregnancy, a 3-hour treatment of estrogen resulted in a significant upregulation of CLU expression in the uterus of ovariectomized mice. These data indicate that Clu is likely regulated by estrogen through ERα and is influenced in a short period. To gain a more comprehensive understanding of the role of Clu in early pregnancy, further experiments are necessary.