Association of maternal thyroid peroxidase antibody during pregnancy with placental morphology and inflammatory and oxidative stress responses

Background Studies suggest that thyroid peroxidase antibody (TPOAb) positivity exposure during pregnancy may contribute to changes in placental morphology and pathophysiology. However, little is known about the association of maternal TPOAb during pregnancy with placental morphology and cytokines. This study focuses on the effect of repeated measurements of maternal TPOAb during pregnancy on the placental morphology and cytokines. Methods Based on Ma’anshan Birth Cohort (MABC) in China, maternal TPOAb levels were retrospectively detected in the first, second and third trimesters. Placental tissues were collected 30 minutes after childbirth, placental morphological indicators were obtained by immediate measurement and formula calculation, and cytokine mRNA expression was detected by real-time quantitative polymerase chain reaction (RT-qPCR) afterward. Generalized linear models and linear mixed models were analyzed for the relationships of maternal TPOAb in the first, second and third trimesters with placental indicators. Results Totally 2274 maternal-fetal pairs were included in the analysis of maternal TPOAb levels and placental morphology, and 2122 pairs were included in that of maternal TPOAb levels and placental cytokines. Maternal TPOAb levels in early pregnancy were negatively associated with placental length, thickness, volume, weight and disc eccentricity, while positively correlated with placental IL-6, TNF-α, CRP, CD68, MCP-1, IL-10, HO-1, HIF-1α and GRP78. In mid-pregnancy, maternal TPOAb levels were negatively correlated with placental length, width and area. In late pregnancy, maternal TPOAb levels were negatively correlated with placental length, area, volume and weight. Repeated measures analysis showed that maternal TPOAb positivity tended to increase placental TNF-α, CD68 and MCP-1 while decreasing placental length, width and area than TPOAb negativity. Repeated measures analysis showed that maternal TPOAb levels were positively correlated with placental IL-6, TNF-α, CD68, MCP-1, IL-10, HO-1, HIF-1α and GRP78, while negatively correlated with placental length, area, volume, weight, and disc eccentricity. Conclusion There may be trimester-specific associations between maternal TPOAb levels and placental morphology and inflammatory and oxidative stress responses. The effect of maternal TPOAb levels on placental morphology is present throughout pregnancy. Early pregnancy may be the critical period for the association between maternal TPOAb levels and placental inflammatory and oxidative stress responses.

Background: Studies suggest that thyroid peroxidase antibody (TPOAb) positivity exposure during pregnancy may contribute to changes in placental morphology and pathophysiology.However, little is known about the association of maternal TPOAb during pregnancy with placental morphology and cytokines.This study focuses on the effect of repeated measurements of maternal TPOAb during pregnancy on the placental morphology and cytokines.
Methods: Based on Ma'anshan Birth Cohort (MABC) in China, maternal TPOAb levels were retrospectively detected in the first, second and third trimesters.Placental tissues were collected 30 minutes after childbirth, placental morphological indicators were obtained by immediate measurement and formula calculation, and cytokine mRNA expression was detected by real-time quantitative polymerase chain reaction (RT-qPCR) afterward.Generalized linear models and linear mixed models were analyzed for the relationships of maternal TPOAb in the first, second and third trimesters with placental indicators.
Results: Totally 2274 maternal-fetal pairs were included in the analysis of maternal TPOAb levels and placental morphology, and 2122 pairs were included in that of maternal TPOAb levels and placental cytokines.Maternal TPOAb levels in early pregnancy were negatively associated with placental length, thickness, volume, weight and disc eccentricity, while positively correlated with placental IL-6, TNF-a, CRP, CD68, MCP-1, IL-10, HO-1, HIF-1a and GRP78.In mid-pregnancy, maternal TPOAb levels were negatively correlated with placental length, width and area.In late pregnancy, maternal TPOAb levels were negatively correlated with placental length, area, volume and weight.Repeated measures analysis showed that maternal TPOAb positivity tended to increase placental TNF-a, CD68 and MCP-1 while decreasing placental length, width and area than TPOAb negativity.Repeated measures analysis showed that maternal TPOAb levels were positively correlated with

Introduction
Thyroid peroxidase (TPO) is the primary enzyme involved in producing thyroid hormones (1).Thyroid peroxidase antibody (TPOAb) acts as a competitive inhibitor of the action of TPO and is responsible for thyroid inflammation (2,3).TPOAb is the most common anti-thyroid autoantibody.TPOAb is a marker of autoimmune thyroid disease (AITD) and would cause thyroid cell damage by activating complement-mediated cytotoxicity and antibody-dependent cell-mediated cytotoxicity (ADCC) (4,5).TPOAb positivity is frequently present in women of reproductive age.The prevalence of TPOAb positivity observed in pregnant women ranges from 2% to 17%, with a higher prevalence in iodine-deficient populations (6).TPOAb positivity enhances the risk of adverse pregnancy outcomes, including developing thyroid disease during pregnancy, miscarriage, preterm birth, placental abruption, premature rupture of membranes and fetal neurodevelopmental delay (7).
Successful pregnancy maintenance depends on immune homeostasis, immune tolerance and relative cytokines levels (2).Studies have demonstrated that the expression of the inflammatory cytokines tumor necrosis factor-a (TNF-a), interleukin-1b (IL-1b), IL-6, IL-8 and monocyte chemoattractant protein-1 (MCP-1) was elevated in the fetal membranes, cervix, amniotic fluid and placenta (8).TPOAb positivity may lead to failure of immune tolerance at the maternal-fetal interface, severely compromising placental-fetal development (4,9).Elevated TPOAb levels may modulate immune activity at the cellular level, leading to preterm birth (10).Animal studies indicated extremely preterm birth and fetal intrauterine growth restriction (IUGR) were associated with placental injury and inflammation, as evidenced by the upregulation of inflammatory cytokine and chemokine genes in placenta, including TNF-a, IL-1b, IL-6, MCP-1, macrophage inflammatory peptide-2 (MIP-2) and keratinocyte-derived chemokine (KC) (11).In addition, TPOAb may diffuse through the placental barrier at all stages of pregnancy, increasing immune responses and affecting placental development and pregnancy progression (9,12).Therefore, it is speculated that maternal TPOAb levels may influence the immune status at the maternal-fetal interface by altering the expression of cytokines in the placenta.
The existence of thyroid autoimmune could indicate a decrease in the ability of the thyroid gland to adapt to the necessary changes related to pregnancy adequately.Hypothyroidism affects fetalplacental development by impairing placental decidualization and vascularization, increasing apoptosis and reducing trophoblast proliferation (13).This may be linked to inflammatory mediators in the placenta.At present, there is only evidence from animal studies suggesting that hypothyroidism affects maternal immune function by interfering with the development of an anti-inflammatory environment, and that maternal hypothyroidism is associated with hypoxia and activation of inflammatory and oxidative stress at the maternal-fetal interface (14,15).Hypothyroidism reduced the expression of placental interferon-g (IFN-g), IL-10 and migration inhibitory factor (MIF) in rats (14), which stimulates the expression of a wide variety of pro-inflammatory cytokines (16).Furthermore, even in the absence of thyroid dysfunction, many studies have linked the presence of TPOAb to adverse maternal-fetal outcomes during pregnancy (9).However, there are no population studies of the association between maternal TPOAb levels and placental cytokines, and it is unclear whether the effect on the placenta is independent or mediated by thyroid hormones.
In the current study, based on a prospective birth cohort study, we had repeatedly measured maternal TPOAb in the first, second and third trimester of pregnancy.We aimed to examine the relationship of maternal TPOAb levels with placental morphology and inflammatory and oxidative stress responses and to identify the potential critical period.

Participants
This study was based on Ma'anshan Birth Cohort (MABC) in China.Pregnant women who had their first antenatal checkup from May 2013 to September 2014 were invited to join the study.The inclusion criteria were as follows: 1) permanent residents in Ma'anshan City; 2) within 14 weeks of gestation; 3) planning to have pregnancy checkups and delivery at Ma'anshan Maternal and Child Health Center; 4) able to understand and complete the questionnaires (22).Further exclusion criteria were set in the current study as 1) twin pregnancy and adverse pregnancy outcomes (spontaneous abortion, therapeutic abortion, ectopic pregnancy, and stillbirth); 2) having a family history of thyroid diseases and/or those who suffered from thyroid conditions (including hypothyroidism, hyperthyroidism, thyroiditis, thyroid tumor/cancer) before/during pregnancy and received treatment; 3) missing data on maternal TPOAb concentrations; 4) missing data on placental morphology or cytokines.
The current study was a retrospective cohort study from prospectively collected data.Women had been prospectively followed up during pregnancy.After delivery, their children were continuously followed up in the cohort.Maternal thyroid function was determined retrospectively during childhood follow up.Women were not aware of their thyroid function from assays of biological samples collected in this study.
All participants provided written informed consent.The study protocol was approved by the Biomedical Ethics Committee of the Anhui Medical University (No. 20131401).

Evaluation of maternal thyroid function
Fasting venous blood from women was collected in the first (before 13 weeks), second (14-27 weeks) and third (28 weeks and beyond) trimester of pregnancy, respectively.After centrifugation, the serum was stored at -80°C.During the children's follow-up period, thyroid-stimulating hormone (TSH), free thyroxine (FT 4 ) and TPOAb levels were determined retrospectively by electrochemical immunoassay (Cobas E411 analyzer; Roche Company, Germany).The detection limit of TPOAb was 5.0 IU/ mL, and the coefficient of variation between the reagent batches was<10% (22).Women with TPOAb positivity were defined as TPOAb >34.0 IU/mL, otherwise were defined as TPOAb negative controls.

Measurement of placental indicators
Placenta tissues were collected 30 minutes after delivery, and the placental length, width and thickness were measured immediately.Placental area, volume, weight and disc eccentricity were calculated using the following formulas (23)(24)(25)(26).

Covariates
Based on the previous literature and directed acyclic graph (Supplementary Figure 1), maternal age, education level, gestational weight gain, monthly income, parity, fetal gender, smoking and drinking during pregnancy were identified as potential confounders.Data on maternal age, education level, monthly income, parity, smoking and drinking during pregnancy were collected questionnaires during recruitment.Information on fetal gender and gestation weight gain was obtained from medical notes.
In addition, TSH and FT 4 levels, gestational diabetes mellitus (GDM), hypertensive disorder complicating pregnancy (HDCP), maternal infection or inflammation during pregnancy, and gestational age at birth were included in the sensitivity analyses, respectively.Maternal infection or inflammation during pregnancy covered bronchitis, influenza, gastroenteritis, cholecystitis, vaginitis, chorioamnionitis, and pelvic inflammatory disease.

Statistical analysis
SPSS 26.0 was adopted for statistical analyses and Graphpad Prism 8.0.2 was used for figure drawing.Statistical significance was declared at P<0.05.
Compared to the differences between the included and excluded, T-test was used for continuous variables, and chisquare test was performed for categorical variables.Since the distributions of TSH, FT 4 , TPOAb concentrations and placental indicators were right-skewed, they were all ln-transformed.Generalized linear models were analyzed for the relationships of maternal TPOAb exposure in three trimesters with placental indicators, respectively.Linear mixed models, a powerful and robust statistical methods for addressing mixed relationships with the same exposure at different times, were fitted to determine the effect of the repeated measurements of maternal TPOAb on placental indicators.
Five sensitivity analyses were conducted.1) TPOAb positivity may be related to an increased risk of clinical or subclinical hypothyroidism (29, 30), and maternal hypothyroidism further affects placental growth and development (9,31).Therefore, thyroid hormone levels may be potential mediators.When analyzing TPOAb in a single trimester, FT 4 and TSH levels in that trimester and TPOAb levels in the other trimesters were further adjusted.FT 4 and TSH data during pregnancy were further adjusted when performing repeated measure analysis.2) Maternal TPOAb may be related to GDM (32).Maternal obesity could influence placental inflammatory status and morphology in human term placenta (33).Maternal GDM was further adjusted.3) Maternal TPOAb may be related to HDCP (34).HDCP may decrease placental blood flow and oxidative stress (35).Maternal HDCP was further adjusted.4) Maternal infection or inflammation may cause placental maladjustment (36,37).Maternal infection or inflammation was further adjusted.5) The effect of thyroid hormones on placental formation and development varies with gestational age (4).Maternal gestational age was further adjusted.

Basic characteristics of participants
A total of 3474 women were recruited as the initial study population.Totally 2274 maternal-fetal pairs were included in the analysis of maternal TPOAb levels and placental morphology, and 2122 pairs were included in that of maternal TPOAb levels and placental cytokines (Figure 1).Baseline characteristics of 2639 included and 835 excluded participants were shown in Table 1.Compared with the excluded participants, the included mothers had a higher level of education, a higher gestational age at birth, a lower TSH concentration in the first trimester, a higher percentage of primiparas, a higher prevalence of HDCP and GDM.

Distribution of maternal TPOAb levels and placental indicators
Maternal TPOAb concentrations during three trimesters, placental morphological indicators and inflammatory and oxidative stress cytokines mRNA expression are presented in Table 2.The geometric means of maternal TPOAb concentrations were 19.35, 13.12 and 15.48 IU/mL in the first, second and third trimester of pregnancy, respectively.The positive rates of TPOAb in early, mid, and late pregnancy were 11.6%, 6.7%, and 7.2%, respectively.The overall rate of maternal TPOAb positivity during pregnancy was 12.6%.

Discussion
In this prospective cohort study, maternal TPOAb levels were found to be negatively related to placental morphology in the first, second and third trimester of pregnancy.Maternal TPOAb positivity tended to decrease placental length, width and area.The  current study also provides longitudinal evidence that maternal exposure to TPOAb during pregnancy may be an important risk factor for maternal-placental immune activation.Significant positive associations between maternal TPOAb levels and placental mRNA expression of IL-6, TNF-a, CRP, CD68, MCP-1, IL-10, HO-1, HIF-1a and GRP78 in early pregnancy were observed.Maternal TPOAb positivity tended to increase TNF-a, CD68 and MCP-1 mRNA expression.Altered placental morphology may reflect its responses to intrauterine stress during early placental development (38).Our results are consistent with previous studies showing lower placental morphological indicators in TPOAb-positive mothers than in control women.Spinillo et al. found that TPOAb-positive women with TSH ≥2.5 mU/L had lower placental volume and area than euthyroid TPOAb-negative women (19).Tissues growing along the length and width of the placenta have different functions and are influenced by different factors.Tissues growing across the width could be involved in maternal-fetal nutrient transport.However, the tissues along the length have different functions but remain to be studied (39,40).Placental thickness is the main dimension of placental growth in late pregnancy and may reflect the vascularization of the chorionic villi.Placental surface area is mainly established before late pregnancy and may indicate the number of spiral arteries supplying the placenta (38,41).
Placental weight can be used as a proxy for fetal metabolic rate.In our study, maternal TPOAb levels were negatively correlated with placental weight.Recently, there have been conflicting studies on the association between TPOAb levels and placental weight.A small retrospective study from Japan showed that placental weight was lower in TPOAb-positive women (20).Männistö et al. found that significantly higher placental weights were observed in TPOAb-positive mothers as well as women with high TSH and low FT 4 levels in early pregnancy based on the Northern Finland Birth Cohort ( 21).An animal study showed that hypothyroidism reduced placental weight in rats (42).Placental weight and disc eccentricity were reported to be directly related to placental stress Repeated measure analysis: Association between maternal TPOAb exposure and placental morphology (n=2274).(A) Maternal TPOAb positive (vs TPOAb negative) increased placental length, width and area.(B) Maternal TPOAb levels throughout pregnancy were negatively associated with placental length, area, volume, weight and disc eccentricity.Adjusted for maternal age, maternal education years, gestational weight gain, monthly income, parity, smoking, drinking and fetal gender.(43).This suggests maternal TPOAb levels may directly affect placentation and impair placental growth and development.
The negative correlation between maternal TPOAb levels and placental morphology in our study was present in all three trimesters, suggesting that maternal TPOAb levels during pregnancy may be an essential risk factor for placental development.Variation in placental morphology is common, but this variation must be limited to a specific range to ensure adequate placental function.The greater the degree of variability in placental morphology, the more severe the decline in function.
Abnormal expression of placental inflammatory and oxidative stress cytokines may respond to high levels of maternal TPOAb.In this study, rising maternal TPOAb levels correlated with an upward trend in inflammatory cytokines.Early pregnancy is the critical period for this association.Increased numbers of Th1 and decreased frequencies of Th2 in the endometrial leukocyte of women with autoimmune thyroid disease would lead to hypersecretion of IFN-g and reduced production of IL-4 and IL-10 (44).Related studies have also shown that the ratio of TNF-a/IL-10 produced by CD3+/CD4+ cells is significantly higher in women with AITD compared to normal controls (45).This suggests that the maternal TPOAb levels moderately aggravate placental inflammatory status and interfere with placental immune tolerance.
There is a unique immunologic state during successful pregnancy, the balance between the anti-inflammatory and proinflammatory environments changes to facilitate the establishment and maintenance of immune tolerance and promote placental development, but excessive inflammation can affect placental function (2,46).Abnormal expression of placental cytokines may lead to insufficiency and increase the risk of adverse pregnancy outcomes, such as pre-eclampsia (PE) and preterm birth (47,48).Ma et al. found higher expression of placental pro-inflammatory cytokines IL-1b, IL-6 and MCP-1 in pregnant women with PE (49).Animal studies have shown that exacerbated levels of IL-6 and TNF-a contribute to placental dysfunction (50).Excessive TNF-a in the placenta may impair trophoblast fusion and hormone production and promote apoptosis (51,52).CD68, primarily localized to lysosomes and endosomes, could be significantly upregulated in macrophages in response to inflammatory stimuli (53).Kim et al. found that maternal CRP was deposited in the human placenta, and its elevated level was related to chorioamnionitis, PE and preterm birth (54).
The maintenance of immune tolerance during pregnancy depends on the expression of anti-inflammatory cytokines, which is critical to prevent fetal immune rejection (55).In our study, TPOAb levels were found to be positively correlated with placental IL-10 mRNA expression in early pregnancy.IL-10 suppresses the production of many pro-inflammatory cytokines, such as IL-6 and TNF-a.IL-10 promotes trophoblast differentiation, inhibits trophoblast invasion and indirectly stimulates angiogenesis (46).It is suggested that increased IL-10 expression may be involved in regulating the adverse effects of inflammation and oxidative stress, which may be a protective mechanism of the placenta against high TPOAb levels.
This study firstly reported that high levels of TPOAb may be associated with elevated placental oxidative stress cytokines.An animal study had shown that hypothyroidism decreased the expression of HO-1, GRP78 and C/EBP homologous protein (CHOP) genes/proteins in the mid-gestation while increased the expression of HIF-1a, CHOP and nuclear factor erythroid 2-related factor 2 (NRF2) genes/proteins (15).HO-1 is a crucial cytokine for immune tolerance and promotes the establishment of an antiinflammatory environment (56).HO-1 expression is induced under hypoxic conditions and modulates IL-10 signaling (57).HIF-1a is a key regulator of the cellular response to the hypoxic environment and plays a critical role in regulating trophoblast differentiation and invasion and in spiral artery remodeling (58).CHOP and GRP78 are known endoplasmic reticulum stress markers, and their elevation promotes trophoblast fusion, syncytialization and invasion (59-61).Spinillo et al. found that in women with positive TPOAb in early pregnancy, the presence of multiple placental pathological features suggested placental hypoxic/ischemic injury (38,62).This suggests that high levels of TPOAb may activate oxidative stress at the maternal-fetal interface.
This study has several strengths.To the best of our knowledge, this is the first study to systematically investigate the potential effect of maternal TPOAb levels in the first, second and third trimester of pregnancy on placental morphology and inflammatory and oxidative stress responses based on a prospective birth cohort study.All information about exposures, outcomes and potential  Some limitations must be recognized.Firstly, only data on the expression of placental transcriptional biomarkers are available in this study, which only indirectly reflects protein translation.But mRNA assay is more feasible than protein levels in a large sample studies measuring multiple cytokines.Secondly, human placenta samples can usually be collected after childbirth.Dynamic placental monitoring can better reflect placental temporal and spatial characteristics, which could be observed by antenatal ultrasound scan.However, in the current study, we did not have data on maternal ultrasound scan.Thirdly, information on local basic iodine levels needed to be in this study, as it would be an important factor related to individual thyroid function.However, Ma'anshan City has yet to be reported as an iodine area.Finally, although many potential confounders were considered in the current study, residual confounding factors could not be ruled out, such as maternal exposure to selenium (64), heavy metals (11,65), and environmental endocrine disruptors (66).

Conclusions
In conclusion, there may be trimester-specific associations between maternal TPOAb levels and placental morphology and inflammatory and oxidative stress responses.The effect of maternal TPOAb levels on placental morphology is present throughout pregnancy.Early pregnancy may be the critical period for the association between maternal TPOAb levels and placental inflammatory and oxidative stress responses.This study provides evidence for the potentially independent effect of maternal TPOAb on placental morphology and function, which act as the pathway to pregnancy outcomes.Close monitoring of women's TPOAb levels during pregnancy is believable to be important for placental development and subsequent fetal health.

TABLE 1 Basic
Characteristics of Included and Excluded Participants (n=3474).

TABLE 2
Distribution of maternal TPOAb concentrations, placental morphological indicators, inflammatory and oxidative stress cytokines in the participants.