CircSTAM inhibits migration and invasion of trophoblast cells by regulating miR-148a-5p/PTEN axis

Background The mechanisms underlying the pathogenesis of preeclampsia (PE) remains unclear. Exploring the molecular players in PE progression can provide insights into targeted therapy. Methods The expression levels of circSTAM in placental chorionic tissues of PE patients and normal pregnant women were compared by RT-qPCR. CircSTAM was knocked down by small interfering RNA to investigate its role in migration, invasion and epithelial-mesenchymal transformation (EMT) of trophoblast HTR-8/SVneo cells. The downstream target of circSTAM was predicted using online bioinformatics resources, and their molecular interaction was examined by luciferase reporter assay. Results CircSTAM was upregulated in PE placenta tissues in comparison to normal placental tissues. CircSTAM knockdown significantly enhanced cellular invasion, migration, as well as EMT. Mir-148a-5p was identified as a target of circSTAM to regulate cell migration and invasion. Mir-148a-5p negatively regulated PTEN expression in trophoblast HTR-8 /SVneo cells. Conclusion In summary, circSTAM upregulation in PE trophoblasts promoted the invasion, migration and EMT. CircSTAM may modulate trophoblast phenotype by impinging on mir-148a-5p/PTEN axis. These data provided novel insights into the pathogenesis of PE.


Introduction
Preeclampsia (PE), also called prenatal eclampsia, is a pregnancy-associated condition characterized by gestational hypertension, which frequently initiates at 20 weeks of pregnancy [1]. It is recognized as a major cause of mortality and morbidity in pregnant women, with an incidence at 2-8% of all pregnancies worldwide [2,3]. PE is linked with multiorgan disorder and its pathogenesis involves genetic and environmental factors, such as abnormal immune responses [4], oxidative stress [5], placental deformation and dysfunction, and inflammatory reactions [6,7]. However, molecular mechanisms underlying the pathogenesis of PE have not been fully understood. Currently, there is no effective treatment strategy for PE, and termination of pregnancy is the only measure for pregnant women with PE diagnosis.
Increasing number of studies have shown the involvement of non-coding (nc) RNAs in the pathogenesis of pregnancy disorder [8][9][10]. Circular RNAs (circRNAs) are endogenous ncRNAs characterized by covalently linked circular structure and the absence of polyadenylated tail or 5-3 polarity [11,12]. They are evolutionarily conserved and display tissue-specific expression pattern [13,14]. CircRNAs are reported to function as miRNA sponge, since they can physically bind to miRNAs and suppress the interaction between miRNAs and corresponding target mRNAs [13,15]. Several cir-cRNAs have been reported to be potential PE biomarkers [16,17]. In addition, a recent study has showed that 151 circRNAs were expressed differentially in the blood samples between PE patients and normal controls, and circSTAM was one of them without functional characterization [18]. Understanding potential roles in regulating the trophoblasts (the major players in placental formation) provide more insights into the etiology of PE. We therefore aim to functionally characterize the potential role of circSTAM in regulating the migratory phenotype of trophoblasts.
In the present study, we reported the upregulation of circ-STAM in placental chorionic tissues of PE patients, which suggests that circSTAM upregulation may serve as a potential biomarker for PE diagnosis. Moreover, using trophoblast cell line as the in vitro model, we showed that circSTAM targets miR-148a-5p to modulate cellular invasion, migration and EMT of trophoblast cells. We further identified PTEN as a potential target of miR-148a-5p. Together, these findings suggest that circSTAM may contribute to the pathogenesis of PE by regulating the mobility of trophoblasts in the placenta.

Sample collection
The present trial enrolled 45 pregnant females at the First People's Hospital of Wenling, of whom 25 were diagnosed with PE patients and the other 20 were normal controls ( Table 1). All of the subjects were screened strictly for the clinical parameters of PE and for other etiologies [1]. Inclusion criteria: patients with systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg and urine protein ≥ 0.3 g/24 h. Exclusion criteria: patients with severe PE (systolic blood pressure ≥ 160 mmHg or diastolic blood pressure ≥ 110 mmHg, liver and kidney function damage and visual impairment); patients with eclampsia (convulsions and hyperspasmia that cannot be ascribed to the diagnosis criteria of PE); patients with previous history of frequent abortion; patients with previous history of hypertension.
The chorionic tissues of the placenta were collected through abdominal wall puncture or vaginal posterior fornix puncture. The tissues were washed three times with saline and treated with diethyl pyrocarbonate to remove the blood, and then stored in liquid nitrogen instantly until further analysis. Informed consents were acquired from all participants. The experimental procedures were approved by the Medical Research Ethics Committee of the First People's Hospital of Wenling.

Quantitative real-time polymerase chain reaction (qRT-PCR)
TRIzol reagent (Invitrogen, CA, USA) was utilized for collecting total RNAs from tissue samples and cell cultures as described in the manufacturer's instructions. The cDNA synthesis was accomplished with the First Strand cDNA Synthesis Kit (Roche, Swizerland) from RNA sample using random primers. Resulted cDNA was diluted to 40 ng/μL and quantified in a PCR System (BioRad, CA, USA) using SYBR premix EX TAQ II kit (Takara, Dalian, China). The following PCR cycling conditions were used:

Cell migration and invasion assay
The migratory and invasive capabilities of HTR-8/ SVneo cells were assayed by transwell assay. The transwell upper chamber (Corning, Cambridge, USA) was coated with 50 mg/L Matrigel (Gibco, CA, USA, at a 1:8 dilution in PBS) and dried at 37 °C for invasion assay. The upper chamber without Matrigel was used for migration assay. 1 × 10 5 cells were seeded into the upper chamber in serum-free medium and 1 ml of 10% serum-containing medium was added to the lower chamber. Following incubation (24 h for migration assays; 48 h for invasion assays), the membrane was washed with PBS and cotton tips were used to clean the upper membrane surface. Afterwards, ethanol (95%) was used to fix the lower surface-adherent cells. Fixed cells were stained with 0.25% crystal violet (Sigma, Germany) for 20 min. Cells were imaged under an inverted microscope (Olympus, Tokyo, Japan) at 100 × magnification.

Wound healing assay
Cell migration was examined by wound healing assay. The cells were inoculated in 6-well plates (1 × 10 6 cells) until the formation of single cell layer. A scratch was produced in the cell monolayer with a 200-μl-pipette tip. The cells were washed 3 × PBS to remove the floating cells. After refilling with culture medium, cells were incubated at 37 °C for 48 h. Cell images were captured using an inverted light microscope at 100 × magnifications. The relative migration distance is calculated as ratio of would distance at 48 h/ would distance at 0 h.

RNA pull-down assay
In pull-down experiment, circSTAM probe and the control probe labeled with biotin (Sangon Biotech) were transfected into HTR-8/ SVneo cells using Lipofectamine 3000 Reagent (Invitrogen, CA, USA) based on the manufacturer's protocol. After 48 h, cells were harvested using IP lysis buffer (Beyotime, Beijing, China). The lysate was incubated with streptavidin-coated magnetic beads (MagnaBindTM Streptavidin Beads, Thermo Fisher Scientific, CA, USA) for 4 h to precipitate biotin-coupled RNA complexes. The beads were washed three times with ice-cold lysis buffer. TRIzol reagent (Invitrogen, CA, USA) was utilized to perform RNA purification and the sample was analyzed via qRT-PCR.

RNase R and actinomycin D treatment
RNase R (Epicentre Technologies, Beijing, China) and actinomyin D (Sigma, Germany) treatment assay was conducted to analyze the stability of circSTAM and GAPDH mRNA. RNA sample was divided equally into two parts: one for Rnase R digestion (Rnase R group), and the other as control group without RNase R treatment. All the samples were incubated at 37 °C for 15 min, and the relative abundance of circSTAM and GAPDH mRNA was examined by RT-qPCR. RNA stability assay was performed by blocking the transcription using 2 μg/mL actinomycin D. RNA samples were collected at different time points by TRizol reagent. The

Luciferase reporter assay
Luciferase reporter plasmids with WT or mutated (MUT) binding sites were prepared by Shanghai Sangon Biotechnology (Shanghai, China). 48 h following transfection, cells were lysed and the luciferase activity of the lysate was examined using a Renilla-Firefly Dual Luciferase Assay Kit (Thermo Fisher Scientific, CA, USA). Each sample was normalized by dividing the activity of firefly luciferase with the control renilla luciferase.

Statistical analysis
Data were expressed as means ± SD (standard deviation) and analyzed by SPSS version 18.0 and Prism GraphPad 8.0 software. Student's t-test was employed to analyze the significance of difference between two groups, and one-way ANOVA and subsequent Tukey's post hoc test was used for multiple comparisons. All quantification experiments were performed three times and the data were the summary of 3 independent measurements. Differences were regarded as significant when P < 0.05.

CircSTAM is highly expressed in placental tissues of PE patients
We first compared the expressions of circSTAM in the chorionic tissues of the placenta between PE patients and normal controls. CircSTAM was highly expressed in placental chorionic tissue of PE patients (Fig. 1A). Receiver operating characteristic (ROC) curve analysis was performed to show the predictability of circ-STAM for PE patients. Youden's index analysis in conjunction with ROC curve suggests that the maximum value of Youden's index was 0.51, and the AUC (area under the curve) value was 0.824 (P = 0.000, sensitivity = 0.96, and specificity = 0.55), suggesting that circSTAM could be used as a potential diagnostic marker for PE (Fig. 1B). To demonstrate the circularity and stability of circSTAM, we performed RNase R and actinomycin D treatment in RNA samples isolated from HTR-8/SVneo cells. As demonstrated by the qRT-PCR findings, GAPDH mRNA level gradually decreased after actinomycin D treatment, while circSTAM level remained relatively stable (Fig. 1C). Additionally, compared with control group, GAPDH mRNA level was reduced after RNase R treatment, while circSTAM level showed no significant change (Fig. 1D).

CircSTAM knockdown regulates migration, invasion and EMT of HTR-8/SVneo cells
To clarify the potential roles of circSTAM in trophoblasts, siRNA-mediated gene silencing was conducted to reduce the expression level of circSTAM in HTR-8/SVneo trophoblast cells ( Fig. 2A). Upon the silencing of circSTAM, trophoblast cells showed increased migration ability (Fig. 2B). Transwell assay further illustrated that knockdown of circSTAM enhanced the migratory (Fig. 2C) and invasive (Fig. 2D) capabilities. As shown in Fig. 2E, the knockdown of circ-STAM caused the increase of N-cadherin and vimentin expression while E-cadherin expression was downregulated, suggesting the role of circSTAM controlling epithelial and mesenchymal transition.

Discussion
CircRNAs have received considerable research attention in cancer biology and in the fields of other pathological conditions. CircRNAs are involved in the regulation of cellular activity mainly through physically adsorbing miRNAs [14]. Their closed-loop structure is resistant to RNase digestion, inhibitor) were determined through WB. * P < 0.05; **P < 0.01; ***P < 0.001 rendering circRNAs reliable markers for diagnostic purpose [19]. In this study, we showed that circSTAM was highly resistant to RNase R digestion, and was stable after the blockage of transcription, which verified its identity as a circRNA.
Several circRNAs have been found to be differentially expressed in PE patients, which have been proposed as potential markers for diagnosis or targets for treatment [20,21]. We reported an elevation of circSTAM expression in the placental chorionic tissues of PE patients, as well as a crucial role of circSTAM in regulating the migratory phenotype of HTR-8/SVneo cells. Crosstalk between circRNA and miRNA is a conventional model for competing endogenous RNA regulatory networks. CircRNAs harbor binding sites for miRNAs and are capable of isolating miRNAs from their regulatory function on mRNAs [22]. Our present study showed that circSTAM suppresses the invasion, migration and EMT of HTR-8/SVneo cells by serving as a mir-148a-5p sponge. The proliferation, invasion, differentiation, migration and apoptosis of trophoblast cells are the basic elements of placenta formation and embryonic development [6,7]. Insufficient trophoblast proliferation is a central cause of PE at early stage. Therefore, our data suggest that circSTAM upregulation in PE patients may impinge on the migratory phenotype trophoblast cells, which may underlie the development of PE.
Mir-148a has previously been reported to be involved in the development of multiple types of cancer, including gastric, hepatic, breast and pulmonary cancers [23]. Furthermore, mir-148a has been described as an integral part of the regulatory circuit of NF-κB pathway, which is activated in Hodgkin and Reed/Sternberg (HRS) cells [24]. However, there is currently no report regarding its role in regulating trophoblast cells. In this study, we found the decreased expression of mir-148a-5p in placental tissues of PE patients. We also reported that mir-148a-5p mediates the role of circSTAM in regulating migration and invasion of HTR-8/SVneo cells. In addition, we also demonstrated that circSTAM could regulate PTEN level by sponging mir-148a-5p. It has been previously reported that PTEN is involved in the molecular pathogenesis of PE development [25]. Specifically, circRNA VRK1 promotes PE progression via sponging miR-221-3P to regulate PTEN/Akt signaling. In addition, long non-coding RNA (lncRNA) LINC01347 also regulates trophoblast migration through the miR-101-3p/PTEN/AKT axis [26]. Therefore, different circRNAs or lncRNAs may connect with different miRNA but regulate similar signaling pathway in PE, such as the regulation of PTEN. On the other hand, it has been reported that metformin shows therapeutic effect for PE by modulating the miR-148a-5p/P28 axis [27]. These findings and our data altogether imply that circRNAs and miRNAs, each can have multiple targets, form an intricate regulatory network in the regulation of PE progression.
Our study is limited by the fact that the cell model is based on the immortalized trophoblast cell line. The observation is required to be validated in the primary trophoblast cells. Furthermore, the mechanisms underlying circSTAM upregulation in PE patients are unclear. Unveiling the molecular mechanism is critical for the formulation of therapeutic strategy to target circSTAM. Further, the role of PTEN dose-dependent effect on the physiological functions of trophoblasts remains to be investigated. Further studies are also needed to consolidate the role of circSTAM in animal model of PE.

Conclusion
Our study showed an upregulation of circSTAM in placental tissues of PE patients and demonstrated its role in controlling migration and EMT in trophoblasts. In addition, PTEN was identified as target of mir-148a-5p. These data suggest that circSTAM/mir-148a-5p/PTEN axis may underlie the pathogenic development of PE, which requires further validation in animal models.
Author contribution L Chen and X Chen mainly participated in literature search, study design, writing and critical revision. J Yan, H Zhang and Jia Xu mainly participated in data collection, data analysis and data interpretation. All authors read and approved the final manuscript.

Data availability
The data in the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate Informed consents were acquired from all participants, while the protocol approval 272 was obtained from the Ethics Committee of The First People's Hospital of Wenling.

Competing interests The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.