Dysfunction of Tregs contributes to FGR pathogenesis via regulating Smads signalling pathway

Abstract Fetal growth restriction (FGR) is ranked number two of most common complication of abnormal pregnancy worldwide. The pathogenesis of FGR is complicated due to multiple aetiologies and the exact mechanism for FGR development is currently unknown. T regulatory cells (Tregs) are proven to play central roles in the maintenance of normal pregnancy. Peripheral blood samples of 102 pregnant human were collected analysed using flow cytometry to identify Tregs. We found that reduced Tregs and down‐regulation of Foxp3 were observed in peripheral blood of FGR patients. In FGR mouse model, we have found that Tregs were not only reduced in spleen but also in placenta. In vitro, Foxp3 and its transcription regulatory signalling molecules, including P‐Smad2, P‐Smad3 and Smad4, were diminished as well. Inhibition on Foxp3 expression was partially reversed by overexpression of Smad2 and Smad4. In FGR patients, Western blot results revealed that Foxp3, P‐Smad2, P‐Smad3 and Smad4 expression was inhibited in placenta. Our preliminary result suggests that maternal‐foetal immune tolerance mediated by Tregs plays an essential role in the development of FGR. The inhibited expression of Foxp3 and down‐regulated Smad2/Smad3/Smad4 signalling pathway were involved in the FGR pathogenesis. Targeting maternal‐foetal immune tolerance through Tregs might represent a novel therapeutic option for FGR.

In FGR mouse model, we have found that Tregs were not only reduced in spleen but also in placenta. In vitro, Foxp3 and its transcription regulatory signalling molecules, including P-Smad2, P-Smad3 and Smad4, were diminished as well. Inhibition on Foxp3 expression was partially reversed by overexpression of Smad2 and Smad4.
In FGR patients, Western blot results revealed that Foxp3, P-Smad2, P-Smad3 and Smad4 expression was inhibited in placenta. Our preliminary result suggests that maternal-foetal immune tolerance mediated by Tregs plays an essential role in the development of FGR. The inhibited expression of Foxp3 and down-regulated Smad2/ Smad3/Smad4 signalling pathway were involved in the FGR pathogenesis. Targeting maternal-foetal immune tolerance through Tregs might represent a novel therapeutic option for FGR.

K E Y W O R D S
fetal growth restriction, Foxp3, Smads signaling pathway, T regulatory cells studies suggest that placenta malfunction, 4 uterus infection, 5 genetic factors and race, 6 as well as environmental pollutions, 7 are all possible causes of FGR. However, the exact mechanism for FGR development is unclear.
Several lines of evidence suggest that maternal immune tolerance to the semiallogenic foetus plays important role in a successful pregnancy and dysregulation of this immune balance is closely related to the development of FGR. 8 Fas ligand (FasL) can bind to its receptor (Fas) on activated immune cells, leading to immune cell apoptosis, which limits the host immune response. During normal pregnancy, maternal immune tolerance is essential for the proper development of the foetus. One mechanism suggests that foetal trophoblast cells express FasL, which subsequently induces apoptosis of activated lymphocytes. Karthikeyan et al indicated that FasL induced by decidual cells was reduced remarkably in FGR pregnancy compared to normal pregnancy, that peripheral blood level of FasL was significantly lower in FGR patients than in normal controls, and that low level FasL was correlated with lower placenta and newborn weights. 9 Placenta expression of human histocompatibility antigen-G mRNA which inhibits natural killer cells and supports survival of semiallogenic foetus was significantly decreased in FGR patients compared to normal controls. 10 Finally, abnormal maternal inflammation can lead to FGR via increased secretion of tumour necrosis factor-a (TNF-a). 11 T regulatory cells (Tregs) are a special subtype of T cells, important for immune regulation. They are marked by CD4, CD25 and Foxp3. Both peripheral blood Tregs and spleen Tregs take a great part in immune homeostasis and immune tolerance during pregnancy. During normal pregnancy, Tregs are increased at the early phase of pregnancy and maintained at constant level until the end of pregnancy. 12 Tregs induce immune suppression via secreting several suppressive cytokines (IL-10, TGF-β), or through direct cell-cell interaction. In the abortion-prone mice, adoptive transfer of pregnancy-induced Tregs can ameliorate the foetal absorption. 13 Female mice lacking conserved non-coding DNA sequence 1 (CNS1) have increased rate of miscarriage due to inability to induce Tregs. 14 It is established that Tregs are involved in miscarriage of unknown cause, but is not known whether Tregs play a role in FGR as well.
In our study, the diminished proportion of Tregs and reduced Foxp3 expression were observed in peripheral blood of FGR patients.
Tregs were not only reduced in spleen but also in placenta in FGR mouse model. In vitro, the expression of Foxp3 and its transcription regulatory signalling molecules, including P-Smad2, P-Smad3 and Smad4, were diminished. Overexpression of Smad2 and Smad4 could partially abrogate the inhibitory effect on Foxp3 expression. Our findings indicate that dysfunction of Tregs may be involved in FGR pathogenesis via regulating Smad2/Smad3/Smad4 signalling pathway.

| Selection of patients
We conducted a prospective study between December 2011 and June 2019 in Nantong University Affiliated Hospital. Informed consent was acquired from patients who visited the obstetrics department prior to sample collection. Pregnancy starting date was determined based on ultrasound measurements between 8-12 gestational weeks. Pregnancies with foetuses having congenital or chromosomal abnormalities or multiple pregnancies were excluded.
According to the American Academy of Family Physicians (AAFP) growth curves, 15 patients were divided into the following groups: 1.
Term (＞37 weeks) and appropriate for gestational age (AGA ＞ 10th percentile); 2. Term and small for gestational age (SGA ≤ 10th percentile). Both maternal serum and placentas were gathered during delivery. Demographics information about the patients was recorded in detail as following (Table 1).

| Ethics Statement
The prospective study protocol and all procedures were approved by

| RNA Isolation and qRT-PCR
Total RNA was isolated utilizing TRIzol reagent (Thermo Fisher Scientific), and cDNA was synthesized utilizing the iScript cDNA synthesis kit (Bio-Rad) according to the instructions. Primers

| Flow cytometric analysis
Maternal placenta samples were gathered, and single-cell suspensions were prepared according to Tang

| Isolation of Tregs and Proliferation assay
CD4 + CD25 + T cells were isolated from the splenocytes according to the manufacturer's instructions. Tregs, whose purity was between 96% and 98%, was used to do subsequent experiments.
CD4 + CD25 − T cells isolated from naive mice were stimulated with 1 µg/ml anti-CD3 mAb in the presence of CD4 + CD25 + T cells isolated from PBS-injected and ZIKV-injected pregnant mice and cultured for 72 hours. ELISA 5-bromo-2-deoxyuridine (BrdU) kit (Roche Applied Science, Mannheim, Germany) was utilized to assay the proliferation.

| Western Blot
Proteins were separated and transferred to PVDF membranes at 300 mA for 2 hours. After blocking, the membranes were then incubated with anti-P-Smad2, P-Smad3, Smad2, Smad3, Smad4

| Immunohistochemistry analysis
Placenta tissue sections from normal pregnancy and FGR patients were deparaffinized and rehydrated through graded alcohols.
Endogenous peroxidase activity was blocked by incubation in 3% Nuclei were visualized with haematoxylin on each slide, which imaged by Olympus BX40 with CellSens Dimension software.

| Data Statistical Analysis
All data statistical analyses were calculated with SPSS20 statistic software and STATA 12.0. Data were showed as mean ± SD. t test was performed to identify the differences between individual treatment and control groups. The results of multiple group comparisons were analysed with ANOVA.

| FGR pregnancy was associated with reduced number of Tregs and down-regulation of Foxp3 in peripheral blood
Maternal Tregs play a dominant role in maintaining immune tolerance during pregnancy. 19 During normal pregnancy, maternal Tregs cells with foetal specificity accumulate, while the number of such Treg cells are reduced during unexplained infertility, miscarriage and preeclampsia. 20 To determine whether Tregs play a role in FGR pregnancy as well, we collected peripheral blood samples from 62 normal pregnancy and 40 from FGR pregnancy for flow cytometer analysis. The proportion of CD4 + /CD25 + /Foxp3 + lymphocytes among CD4 + lymphocytes in FGR pregnancy (7.38%) was significantly lower than in normal pregnancy (16.5%) (P = .011) ( Figure 1A). In addition, Foxp3 expression in peripheral blood was significantly lower in FGR pregnancy than in normal pregnancy ( Figure 1B). These data suggest that the deficit in Treg cell number may be involved in FGR pathogenesis.

| Diminished number and reduced function of Tregs were observed in FGR mouse model
To further explore the correlation between Tregs dysfunction and FGR pathogenesis, we establish FGR mouse model by ZIKV infection. Mice pregnant at E10.5 were injected with ZIKV and followed until E18.5. ZIKV-infected mice had significantly lower bodyweight starting at E15.5. Further, we showed that both size and weight of foetuses from ZIKV-infected mice were significantly reduced compared to normal mice (Figure 2A-2C). Thus, we successfully established FGR mouse model. Then, we determined the number of Tregs in the spleen of FGR pregnancy mice by flow cytometer.
As shown in Figure 2D, the percentage of CD4 + /CD25 + /Foxp3 + cells in the spleen of FGR mice (8.45%) was significantly lower than in the spleen of control group (15.38%). The percentage of CD4 + / CD25 + /Foxp3 + cells in the placenta of FGR mice (2.66%) was also significantly lower than in the placenta of control group (6.85%) ( Figure 2E). To further assess the function of Tregs, we isolated the primary Tregs (CD4 + CD25 + T cells) from FGR mouse model. ZIKV infection resulted in the attenuation of the inhibitory capacity of CD4 + CD25 + T cells ( Figure 2F). These data indicated the diminished number and reduced function of Tregs may be correlated with FGR pathogenesis.

| ZIKV infection lowered Foxp3 expression in EL-4 cells
To determine the mechanism of how ZIKV reduced Tregs, we directly infected murine lymphoblast cell line EL-4. We showed that ZIKV infection could directly down-regulate Foxp3 protein expression ( Figure 3A). As Smad signalling pathway is directly involved in Foxp3 expression induction and maintenance, 21 we determined the expression of Smad2, Smad3 and Smad4, as well as P-Smad2 and P-Smad3.
As shown in Figure 3B, ZIKV infection did not affect Smad2 and Smad3 expression, but significantly down-regulated Smad4 expression as well as P-Smad2 and P-Smad3 (P < .05). These data suggest that in ZIKV-induced FGR animal model, Smad signalling pathway might directly participate in FGR pathogenesis.

| Foxp3 expression was inhibited through Smad2/3/4 signalling pathway
To further elucidate the mechanism of Smad2/3/4 signalling pathway in FGR pathogenesis, we overexpressed Smad2 and Smad4 and determined their effects on Smad2 and Smad4 expression as well as their downstream genes. We showed that overexpression of

| Smad2/3/4 signalling pathway participated in FGR pathogenesis
In order to further confirm the involvement of Smad2/3/4 signalling pathway in FGR pathogenesis, we determined the expression levels of relevant proteins in the placenta of FGR patient. Analysis of Western blot identified that Foxp3 was remarkably downregulated, as well as P-Smad2, P-Smad3 and Smad4 ( Figure 5A).
Immunohistochemistry analysis further confirmed down-regulation of Foxp3 and Smad4 in the placenta from FRG patients ( Figure 5B).
Our data suggested that Smad2/3/4 signalling pathway participated in FGR pathogenesis.  22 Tregs numbers were significantly lower in FGR cord blood than in normal cord blood, accompanied with suppressed expression of Foxp3. 15 Preeclampsia is the major and cell proliferation via reduction of arginine levels. 27 Enhanced NKT cells in recurrent abortion or implantation failure lead to pregnancy loss can be ameliorated with intravenous immunoglobulin (IVIG) treatment, and achieve successful pregnancy. 28 Future studies should investigate their roles in FGR as well as interaction among these different types of immune suppressive cells.

| D ISCUSS I ON
Foxp3 is a specific transcription factor of Tregs, which regulate

ACK N OWLED G M ENT
This study was supported by Science and Technology Innovation and Demonstration Promotion Project, MS22018001, Nantong, Jiangsu, China.

CO N FLI C T S O F I NTE R E S T
The authors declared that they have no competing interests.

AUTH O R CO NTR I B UTI O N S
YZX and WLG designed the study; YZX and MS collected the tissue samples; ZHW performed the mouse model; QQL and XYX collected clinical data and participated cells experiments and flow cytometric analysis; STG and WDP did Tregs isolation experiments and proliferation assay; YZX drafted the manuscript; and WLG supervised the study. All authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analysed during this study are included in this article. F I G U R E 5 Foxp3 and Smad4 protein expression in placenta from FGR pregnancy. A, Western blot analysis of Foxp3 and Smad4 protein expression as well as phosphorylation of Smad2 and Smad 3 protein in placenta from normal and FGR pregnancy. B, Immunohistochemistry analysis of Foxp3 and Smad4 protein expression in placenta from normal and FGR pregnancy. *P < .05