Dysfunction of Shh signaling impaired trophoblast motility and autophagy is involved in poor placentation of recurrent miscarriage


 Background: In early pregnancy, the placenta anchors the conceptus and supports embryonic development and survival. This study aimed to investigate the underlying functions of Shh signaling on recurrent miscarriage, an serious disorder of pregnancy. Methods: Immunofluorescence and immunohistochemistry were used to detect protein expression and its location in placental tissues. Quantitative real-time RT-PCR and Western blot analysis were performed to examine mRNA and protein levels, respectively. Lentiviruses expressing short hairpin RNA were used to knock down the target genes. Cell invasion and migration were performed by with or without Matrigel-coated transwell, respectively. Primary trophoblast migration was performed by villous explant assay. RNA-sequence was used to investigate the genes transcription profile. CCK-8 assay was used to evaluate cell viability. Flow cytometry was used to evaluate cell apoptosis. Results: Our results showed that Shh and Gli2 were mainly located in cytotrophoblasts (CTBs), Ptch was mainly located in syncytiotrophoblasts (STBs), while Smo and Gli3 were expressed in both CTBs and STBs. Compared to the gestational age-matched normal human placenta, the expression of Shh was significantly decreased in recurrent miscarriage. Furthermore, inhibition of Shh signaling impaired motility of JAR cells via regulating the expression of Gli2 and Gli3 to decrease AKT Ser473 phosphorylation, elevate E-cadherin and VEGFA. Intriguingly, inhibition of Shh signaling also enhanced autophagy and autolysosome. Additionally, knockdown BECN1 reversed the effect of Gant61 on motility inhibition. Conclusion: Our results indicated that dysfunction of Shh signaling impaired trophoblast motility, angiogenesis and activated autophagy in villous trophoblast, which would contribute to the pathophysiology of recurrent miscarriage.

miscarriage is a complex context that involving in multiply signalings and steps dysfunctions on development of trophoblast and placentation.
Sonic hedgehog (Shh) together with Indian hedgehog (Ihh) and Desert hedgehog (Dhh) consist of the ligands of Hedgehog(Hh) signaling pathway, which plays pivotal roles in regulating cell proliferation, cell differentiation, organogenesis, and development, even involved in tumorigenesis and progression [22,23]. Previous study had summarized and indicated that Hh signaling played important roles in regulating hematopoiesis, vasculogenesis and angiogenesis during embryogenesis and development [24,25]. Moreover, Shh might crosstalk with TGF/SMAD signaling pathway to promoted G-CSF mobilized human CD34 + cell migration, proliferation and then differentiation into vascular cells during embryonic vascular development [26]. Our previous study found that Shh together with Gli2/3 was required for the proper development of placental and pregnant maintenance [27]. In addition, we also found that Hh signaling through GLI1/2 promoted EMT process of human trophoblast cells via transcriptional suppression of CDH1 gene [28]. However, the relationship between Shh and recurrent miscarriage, as well as the underlying functions of Shh on recurrent miscarriage, have not been investigated.
Autophagy is a lysosomal degradative process, which contributes to maintain development and homeostasis [29][30][31]. Recent studies have reported that Hh signaling can regulate autophagy in various cells, while functions of Hh signaling on autophagy were controversial. Inhibition of the Hh signaling can induce autophagy in Drosophila, HeLa cells, and human hepatocellular carcinoma cells [32,33]. However, the Hh signaling acts as a positive regulator in autophagy of vascular smooth cells and hippocampal neurons [34,35]. Additionally, the combined inhibition of Hh signaling and autophagy can overcome chronic myeloid leukemia drug resistantance, which was associated with PARP cleavage, CASP3 and CASP9 cleavage, and the BCR-ABL oncoprotein [36]. Other studies showed that the autophagy was highly activated in EVTs and placentas, and its disruption associated with preeclampsia and intrauterine growth restriction [37][38][39][40][41]. However, the effect of autophagy on recurrent miscarriage still remains largely elusive.
In the present study, we investigated the potential role of Shh/Gli signaling and autophagy in recurrent miscarriage. Our results showed that Shh signaling was attenuated in the placenta of recurrent miscarriage patients, dysfunction of Shh/Gli impaired trophoblast migration and angiogenesis. Moreover, inhibition of Shh signaling enhanced autophagy, while inhibition of autophagy would reverse Gant-6 induced inhibition of trophoblast motility. Thus, these findings gave rise to an important implication for the pathological role of Shh signaling and autophagy in recurrent miscarriage and might provide potential targets for recurrent miscarriage therapy.

Material And Method 3.1 Preparation of placental tissues
Placental villi tissues were isolated from normal pregnant women who were voluntary to abortion by themselves (n = 10) between the age 27 and 40 years and the control group had an abortion at 49-82 days of gestation. Recurrent miscarriage patients (n = 10) between the ages of 24 and 37 years(mean age 32.2 ± 5.45 years) at 33 and 66 days of gestation (mean gestational age 44.2 ± 12.83 days) were recruited for this study. None of them had any risk factors such as genetic abnormalities (neither themselves nor their husbands), uterine malformation, thyroid dysfunction, anti-phospholipid antibody syndrome. All samples were stored at -80 °C for further use or stored in 4% formaldehyde at room temperature overnight for immunohistochemistry analysis. The study protocol was approved by the ethics committeeof the Sir Run Run Shaw Hospital, Zhejiang University School of Medicine.

First-trimester villous explant and Cel line culture
Villous explant cultures were captured from first-trimester(6-9 weeks of gestation) as described previously [62]. In brief, 8-10 explants(2-3 mm) from tips of the placental villi were dissected and explanted in 24-well culture dish pre-coated with 5 mg/ml Matrigel for 30 min at 37 °C. Allowing the anchorage for 2 to 4 hours, tissues were carefully covered with 0.5 ml serum-free DMEM/F12 medium and incubated for next 24

Chemicals, Reagents and Oligonucleotides
Cyclopamine (Tocris,1623, USA), recombinant human Shh (R&D,1845-SH), chloroquine (Sigma, C6628), GANT61 (Selleck Chemicals,S8075), Lipofectamine 3000 (Thermo Fisher Scientific, 3.4 Cell counting kit-8 assay JAR Cells were seeded into 96-well plates at 1000 cells/well and cultured for 24 hours, then culture medium was replaced with fresh complete growth medium plus a final concentration of 5 µM cyclopamine (Tocris,1623, USA) or 0.5 µg/ml recombinant human Shh (R&D,1845-SH), complete growth medium plus equal volume alcohol was set as negative control. The maintenance medium was refreshed every two days with the same formulation as previous treatment. After the indicated timepoint treatment, 10 µl cell counting kit-8 (Dojindo, Japan) solution was added to each well and then incubated for 1 hour, cellular vitality was measured by SpectraMax M5 (Molecular Devices, USA) according to the manufacturer's instructions.
3.5 Flow cytometry for Apoptosis assay JAR Cells were seeded into 60 mm disks at 5 10 5 cells/well and cultured for 24 hours, culture medium was then replaced with fresh complete growth medium plus a final concentration of 5 µM cyclopamine or 0.5 µg/ml recombinant human Shh and culture for the next 24 hours, complete growth medium plus equal volume alcohol was set as negative control. At the end point of treatment, cells were gently detached by TrypLE™ Express Enzyme (1 ) (Gibco by Life Technologies, USA). Cells were washed and resuspended with 100 µl binding buffer, 5 µl PE-conjugated Annexin V and 5 µl 7-AAD were immediately added and incubated for 15 minutes in the dark at room temperature. After incubation, all samples were added another 400 µl binding buffer and immediately quantified using the FACSCanto II flow cytometer (BD Biosciences, USA) according to the manufacturer's instructions.

Cell migration and invasion assay
For migration assay, JAR Cells were resuspended with 100 µl basic RPMI-1640 medium and seeded into 24-well transwell polystyrene plates (COSTAR, USA) at 3 10 5 cells/well. For invasion assay, JAR Cells were seeded into 24-well transwell polystyrene plates which pre-coated matrigel (BD, USA).
After transfecting with various shRNA (scramble shRNA, shGli2, shGli3 or shBeclin1), transfected cells were divided into three groups and add another 100 µl basic RPMI-1640 medium plus a final concentration of 5 µM cyclopamine or 0.5 µg/ml recombinant human Shh to each well, basic RPMI-1640 medium plus equal volume alcohol was set as a negative control. The lower chambers were loaded fresh 600 µl complete grown medium into each well. After culture for 24 hours, cells within inner insert were wiped by cotton sticks, migrated cells were fixed with 75% alcohol for 10 minutes at room temperature and then stained by 0.1% crystal violet. After washing with PBS twice, migrated cells were captured by microscope and each image was analyzed by Image J software.

Immunohistochemistry
For paraffin-embedded sections, samples were normally deparaffinized and rehydrated in xylene and a graded series of ethanol, antigen retrieval of samples were boiled in citric acid buffer (pH 6.0), and then incubated with 3% H 2 O 2 to suppress the activity of endogenous peroxidase. Samples were further blocked in 10% normal serum with 1% BSA in TBS for 1 hour at room temperature, and then incubated with primary antibodies overnight at 4 °C. Then, the sections were stained with the HRPconjugated secondary antibodies for 30 min, followed with incubation of diaminobenzidine (DAB) solution, and counterstained with hematoxylin. Primary antibodies used in immunohistochemistry were as following: anti-Sonic Hedgehog (Abcam, ab53281), anti-Cytokeratin 7 (Gene Tech, GM701802), anti-VEGFA (Abcam, ab51745), CD31 (Abcam, ab134168).
For frozen sections or cells grown on glass coverslips, samples were incubated in 0.3% Triton X-100 for 10 min, blocked in 5% goat serum for 1 hour at room temperature, and incubated with primary antibodies overnight at 4 °C. Then, the sections were incubated with fluorescent secondary antibodies for 1 hour at room temperature in the dark, mounted with a drop of mounting medium containing DAPI. The imagines were captured by Olympus BX53 fluorescence microscope, and then analyzed by

Analysis of GFP-mCherry-LC3 puncta
For autophagosome maturation assays, JAR cells were transfected with GFP-mCherry-LC3 adenovirus vector (5 × 10 9 pfu/ml) more than 24 h and treated with or without Hh antagonist chloroquine (CQ) for 24 hours. As the GFP fluorescence is diminished in the acidic autolysosomes while mCherry can be visualized. GFP-mCherry-LC3 puncta were captured by Olympus BX53 fluorescence microscope, and then quantified ImageJ software as described [64].

Statistical analysis
All the numerous data are expressed as the mean ± S.E.M, and were analyzed by Student's t-test or one-way ANOVA (SPSS 13.0J software; SPSS, Inc., Chicago, IL, USA). Statistical significance was assessed at P 0.05. Experiments were independently triplicated, and results were qualitatively identical. Representative experiments are shown.

Shh signaling was attenuated in recurrent miscarriage
We first investigated the expression and cellular localization of Shh signaling core members in villous tissue from first-trimester placentas. 11 β-HSD2 served as a marker of STB layer and played important roles in fetus development [42]. Our results showed that both Shh and Smo were expressed in STB layer and CTB layer, but Shh mainly located in CTB layer. Otherwise, Ptch mainly located in STB layer (Fig. 1A). Compared to normal human placental villi (the healthy control), Shh was decreased in recurrent miscarriage patients' villous tissues (Fig. 1B, 1C). This result was also confirmed by immunoblotted Shh in villous tissues from the healthy control and recurrent miscarriage patients (Fig. 1D). Moreover, CK7, the marker of trophoblast, was decreased in recurrent miscarriage patients' villous tissues (Fig. 1E, 1F), and the thickness of CTB layer were significantly decreased in recurrent miscarriage patients while compared to the healthy control (Fig. 1E, 1G). Thus, Shh signaling was impaired in recurrent miscarriage patients' villous tissues.

Attenuating Shh signaling inhibited trophoblast motility and placental angiogenesis
To study the role of Shh signaling in trophoblast motility, matrigel-coated cell invasion and transwell cell migration models were performed. Our results showed that Smo antagonist cyclopamine (Cyc) significantly decreased 2-fold migration and 1.67-fold invasion in JAR cells while compared to control group (p 0.01 and p 0.05, respectively) ( Fig. 2A-2C). We then investigated the effect of cyclopamine or recombinant human Shh (rShh) on JAR cells' viability. Our results showed that 5 µM cyclopamine 0.5 µg/ml rShh had no effect on JAR cells' proliferation or apoptosis while compared to control group ( Figure S1). Besides, cyclopamine significantly decreased outgrowth of invasive extravillous trophoblasts in first-trimester villous explant cultures seeded on collagen I (Fig. 2D, 2E). On the molecular event, cyclopamine significantly decreased phosphorylation of AKT S473 site, but not AKT T308 site (Fig. 2F). VEGF-A is predominantly important in vasculogenesis during placental development. Our results showed that cyclopamine significantly decreased VEGF-A, while significantly elevated E-cadherin expression in JAR cells (Fig. 2G). Moreover, VEGF-A and CD31 expression were significantly decreased in recurrent miscarriage, while VEGF-A was strongly expressed in trophoblast, and CD31 was strongly expressed in the fetal vessels in villous tissue from healthy controls (p 0.001 and p 0.001, respectively) ( Fig. 2H-2K). These results indicated that downregulation of trophoblast motility might account for aberrant placental vessel regression in recurrent miscarriage patients.

Shh signaling regulated motility of JAR cells via Gli2 and Gli3
To identify downstream factors of Shh signaling on regulationin motility of JAR cells, we first investigated expression of Gli in normal human first-trimester placenta tissues. Our results showed that Gli1 was weakly detected in STB layer, while Gli2 was strongly located in CTB layer, Gli3 was strongly expressed in both STB layer and CTB layer (Fig. 3A). We previously study has showed that Hh regulated EMT in JEG3 cells through Gli1 and Gli2 [43]. Herein, we successfully stably downregulated expression of Gli2 and Gli3 in JAR cells for further experiments (Fig. 3B, 3C). Gli2 knockdown significantly inhibited migration and invasion of JAR cells, while Gli3 knockdown alone or recombinant Shh treatment significantly promoted migration and invasion of JAR cells. Moreover, Gli2 knockdown would partly impair recombinant Shh-induced migration and invasion of JAR cells (Fig. 3D-3F).

Transcriptome Profiling of JAR treated with Cyclopamine
To further screem the cyclopamine-regulated target gene in JAR cells, we analyzed the transcriptome profile of JAR cells with or without cyclopamine treatment. RNA-seq data identified 1797 significant changing genes (P ≤ 0.05, q ≤ 0.05, fold change ≥ 2), and 876 genes were up-regulated, 921 genes were down-regulated (Fig. 4A). Gene Set Enrichment Analysis showed that hypoxia, apical junction and EMT as the top three significantly associated pathway, overall enrichment score of which were 0.47, 0.46 and 0.43, respectively. Normalized enrichment score of which were 2.88, 2.51 and 2.51, respectively (nominal p-value ≤ 0.001,false discovery rate [FDR] ≤ 0.001) (Fig. 4B-4D).

Inhibition of Shh signaling induces autophagy and autolysosome
previous studies showed a close relationship between Shh signaling pathway and autophagy in normal development and tumorigenesis [32][33][34]44], we reasoned that this relationship between Shh signaling and autophagy might involve in initiation and progression of recurrent miscarriage. Our results showed that cyclopamine and Gant61 significantly elevated the ratio of LC3-II/LC3-I protein level in a dose-dependent manner, and this accumulation was enhanced by lysosomal inhibitor chloroquine (CQ) treatment (Fig. 5A, 5B). Gant61 increased the number of mCherry + GFP + yellow puncta (indicating co-localization of mCherry and GFP) in both CQ-treated and untreated JAR cells (Fig. 5C, 5D). Besides, the overlap of LC3B and LAMP1 puncta (the indicator for autophagosomelysosome fusion) was increased in Gant61-treated JAR cells (Fig. 5E, 5F).

Downregulation BECN1 rescued the Gant61-induced inhibition of cell motility in JAR cells
To further investigate the potential role of autophagy on Gant61-induced inhibition of cell motility, we used siRNA to knockdown BECN1, the core autophagy molecule, which as a part of PI3K complex and regulated the localization of other autophagy proteins to phagophores [45,46]. BECN1 was successfully knockdown in JAR cells (Fig. 6A). Downregulation BECN1 significantly reversed Gant61induced inhibition of migration and invasion in JAR cells (Fig. 6B-6D). These data showed a critical role of BECN1on regulating Gant61-induced inhibition of JAR cells' motility.

Discussion
Accumulating promising efforts have been made to improve the prognosis and treatment of unexplained recurrent miscarriages [47,48]. However, the causes and pathophysiology of recurrent miscarriage remain largely elusive. Previous studies so far have focused on abnormal trophoblast development and placentation in recurrent miscarriage [13, 19-21, 23, 24]. Hh signaling was essential for hematopoiesis, vasculogenesis and angiogenesis during embryogenesis and development [24,25].
Our previous studies have indicated that Hh signaling played pivotal roles in the development of placental and pregnant maintenance [27,28]. Besides, in the clinic context, early miscarriage patients have several common traits, including deficient trophoblastic invasion, decreasing thickness of CTB layer, deficient myometrial spiral artery remodeling [16,49,50]. Thus, it rationally prompted us to further explore the underlying relationship between the Shh signaling and recurrent miscarriage. In the present study, our results uncovered a crosstalk between Shh signaling and autophagy on regulating trophoblast motility (Fig. 8). Thus this study suggested that dysfunction of Shh signaling inhibited the trophoblast motility through regulating autophagy and VEGFA, which subsequently resulted in recurrent miscarriage.
Our previous studies have indicated that Hh signaling was required for EMT process of human trophoblast cells, the pregnant maintenance, placental development [27,28,[51][52][53]. Thus, in the present study, we firstly investigated the expression of Shh signaling core members in health human placental villi, our results showed that the Shh/Ptch/Smo/Gli2/Gli3 signaling axis preferred to be activated in CTB layer, and Shh protein was decreased in villous tissue of recurrent miscarriage patients while compared to the healthy control. Recently, the other study indicated that both previous miscarriages and single nucleotide polymorphisms (SNPs) rs3738880 in Gli2 were associated with anorectal malformations, though the relationship between miscarriages and SNP rs3738880 in Gli2 remains unknown [54]. Besides, conditionally deleted of Smo in the mouse uterus impairs implantation and subsequent pregnancy loss [55].
During placental development, migration and invasion of trophoblast cells were essential for placental angiogenesis [3,56]. In the molecular events, PAPP-A2 attenuated HTR8/SVneo trophoblast migration and invasion via decreasing expression of Gli1/2, Snail, Slug, N-Cadherin and Vimentin, while increasing expression of E-Cadherin and ZO-1 [56]. Our previous study also showed that Gli1/2 induced expression of Snail, Slug and Twist1, while suppressed expression of E-Cadherin to promote the migration of JEG3 trophoblast cells [28]. Otherwise, The other study had shown that rosiglitazone increased phosphorylation of AKT to promote endothelial cell migration [57]. In the present study, our results further showed Shh-Gli2/Gli3 played an important role in JAR migration and invasion. Besides, our results showed that inhibition of Shh signaling would decrease phosphorylation of AKT while increase expression of E-cadherin in JAR cells. Thus, our results might provide an alternative mechanism that inactivation of Shh signaling attenuated migration and invasion of JAR cells via decreasing phosphorylation of AKT S473 site while increasing expression of E-cadherin. Moreover, It's reported that the inhibition of Hh pathway induced autophagy through downregulating the AKT-MTOR pathway [58]. In the present study, our results showed that there was no significantly change of autophagy-related genes upon cyclopamine treatement by using RNA-sEq. However, inhibition of Shh signaling promoted autophagosome maturation. Besides, our results further showed that inhibiting autophagy by Beclin1 knockdown would reverse Gant61-induced the motility inhibition of JAR cells. It suggested that Shh signaling interplayed with autophagy in regulating trophoblast motility.
Accumulating studies have been reported that Hh signaling and VEGF-A were required for placental angiogenesis [24,25,[59][60][61]. In the present study, our results also showed that inhibition of Shh signaling decreased the expression of VEGF-A in JAR cells. Additionally, we found that VEGF-A and CD31 were downregulated in villous tissue of recurrent miscarriage patients while compared to the health control. Thus, these results suggested poor vascular placentation in recurrent miscarriage patients.
In conclusion, we provided an alternative mechanism of Shh signaling in regulating trophoblast cells motility through AKT and VEGFA, which subsequently played an important role in placentation and vascularization in recurrent miscarriage patients. Besides, we found that Gant61 could induce autophagy and autolysosome in trophoblast cells, and Beclin1 reponded to Gant61-induced the motility inhibition. Taken these together, our results firstly showed that Shh signaling interplayed with autophagy in human placenta, dysfunctions of which would account for initiation and progression of recurrent miscarriage. Therefore, restoration both of Shh signaling and autophagy might be a potential and promising therapeutical strategy to improve vascularization and penetrate maternal spiral arteries to overcome recurrent miscarriage in future.

Conclusion
Our results here showed that dysfunction of Shh signaling impaired trophoblast motility and angiogenesis in villous trophoblast to contribute to the pathophysiology of recurrent miscarriage, otherwise, Activation Shh signaling by recombinant Shh treatment, Gli3 knockdown or autophagy inhibition by BECN1 konckdown would improve trophoblast motility. Thus, our results provided a new promising strategy that combination targets of the Shh pathway and autophagy in recurrent miscarriage therapy.

Ethical Approval and Consent to participate
The study protocol was approved by the ethics committee of the Sir Run Run Shaw Hospital, Zhejiang University School of Medicine. Preparation of Placental villi tissues were acquired consents from normal pregnant women and recurrent miscarriage patients.

Consent for publication
All authors were consent for publication.