Associations between maternal serum concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and pulse pressure in early pregnancy

Xiaofang Liu Hubei Provincial Center for Disease Control and Prevention Wencheng Cao Hubei Provincial Center for Disease Control and Prevention Xiao Liu Hubei Provincial Center for Disease Control and Prevention Yan Zhou Hubei Provincial Center for Disease Control and Prevention Sheng Wen (  wenshenggy@aliyun.com ) Hubei Provincial Center for Disease Control and Prevention https://orcid.org/0000-0002-0700-1730


Introduction
Current guidelines for the diagnosis and management of hypertension have de ned cardiovascular risk by the elevation of systolic blood pressure (SBP) and/or the elevation of diastolic blood pressure (DBP) (Gillebert, 2018). However, the value of pulse pressure (PP) in the clinical assessment of cardiovascular risk has become increasingly evident (Dart AM & Kingwell BA, 2001). PP is de ned as the difference between SBP and DBP, and various studies have veri ed PP as an independent predictor of cardiovascular risk (Çimen et al., 2017).
Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) compounds are ubiquitous in the environment and bioaccumulate and biomagnify through the food chain due to their persistence and lipophilicity (Olsman H et al., 2007). They were shown to induce carcinogenicity, endocrinopathy, neurotoxicity, and immunotoxicity (Yu DZ et al., 2019). A handful of studies suggest that exposure to dioxins may also be associated with an increased risk of chronic diseases, such as hypertension, ischemic heart disease, stroke, diabetes, and hyperlipidemia (Nakamoto et al., 2013;Pelcl T et al., 2018;Uemura et al., 2009;Warner et al., 2013).
Pregnant women are a special group of people who may intake more food of animal origin, which contributes to more than 90% of dioxins body burden (Lignell et al., 2016). Studies undertaken on maternal participants highlighted epidemiological evidence of short-and long-term effects of in utero exposure. These include associations between maternal exposure, diet, and age with hormonal disruptions in children and impacts on estrogenic metabolism (Baba et al., 2018;Cao et al., 2008;Lignell et al., 2016;Miyashita et al., 2018;Nakajima et al., 2017;Papadopoulou et al., 2014). Additionally, investigative studies on fetal exposure to dioxins provide evidence of an association between maternal serum and fetal abortion, birth defects, and low birth weight (Tuyet LTN & Johansson A, 2001). Although higher concentrations of dioxins have been detected in the placenta and umbilical cord blood, venous blood samples during pregnancy are considered the most representative to evaluate maternal or fetal body burdens (Suzuki, 2005;Wang SL et al., 2004). In recent years, numerous studies have supported increasing evidence of adverse effects of PCDD/Fs in pregnant women. These include the populations in Mexico (Rodríguez Dozal S et al., 2011), Japan (Baba et al., 2018;Todaka et al., 2010), Vietnam (Boda et al., 2018), Taiwan (Huang et al., 2017), Beijing (Bao et al., 2020) Moreover, emerging data support the plausible contribution of speci c environmental toxicants to BP, including dioxins and dioxin-like compounds. A number of studies conducted in populations have shown signi cant associations between dioxin-like compounds and increased BP or hypertension, although the associations with individual congeners are not consistent across studies (Cypel YS, Kress AM, Eber SM, Schneiderman AI, & Davey VJ, 2016;Goncharov A, Pavuk M, Foushee HR, & Carpenter DO, 2011;Nakamoto et al., 2013;Uemura et al., 2009). Animal studies show that 2,3,7,8-TCDD exposure can affect the BP of experimental animals, but the effect may be related to the dose or the species of the animals tested (Hermansky SJ, Holcslaw HL, Murray WJ, Markin RS, & Stohs SJ, 1988;Kopf PG, Huwe JK, & Walker MK, 2008;Timothy P. Dalton et al., 2001).
Gestational hypertension during pregnancy is a common obstetric disease, accounting for 5%~10% of all pregnancies and causing approximately 10%~16% of all pregnancy-related deaths (Fenton TR, Lyon AW, & Rose MS, 2011). However, no reports have been found on the relationship between PCDD/F exposure and PP during early pregnancy. In the present study, we addressed this issue by conducting a crosssectional study to assess the risks of dioxin exposure and the association with PP in early pregnancy.

Materials And Methods
Page 4/19

Study design and population
We conducted a birth cohort study (H-YCCS) in Yingcheng, Hubei Province, China, from May 2018 to February 2021 to investigate the associations between dioxin exposure in early pregnancy and adverse health outcomes. Native and healthy pregnant women were recruited, and prenatal examinations were conducted in Yingcheng People's Hospital, China.
A total of 722 pregnant women were recruited, and 605 (83.8%) of them completed the questionnaire. The questionnaire included general demographic characteristics (age, height, weight, ethnicity, education level, monthly household income, occupation, occupational exposure to hazard factors, alcohol consumption, active and passive smoking, reproductive history, drug use, and disease). BP (SBP and DBP) measurements of pregnant women were obtained by an automatic standard sphygmomanometer beginning after the patient had been sitting for 5 min. Finally, 305 (50.4%) women provided su cient blood samples for PCDD/F measurements. Our study was approved by the Ethics Committee of Hubei Center for Disease Control and Prevention.

Blood collection and chemical analysis
Fasting venous blood samples were collected from the participants by nurses. The whole blood sample was centrifuged immediately, and the serum was segregated and preserved at − 40°C until further analysis. PCDD/Fs in serum were extracted and determined as previously described (Yu DZ et al., 2019).
Brie y, approximately 2 mL of serum was mixed with diatomite and then spiked with 13 C-labeled internal standards (EPA-1613LCS, Wellington Laboratories Inc., Canada). The mixture was extracted by an accelerated solvent extractor (ASE-914, BÜCHI, Switzerland) at 130°C and 100 bar. The extract was concentrated to near one milliliter and then puri ed using an acid silica gel column and carbon column under a pressure of 0.1 MPa (F12, CAPE Technologies, U.S.). Then, the PCDD/F fraction was collected and concentrated to near dryness and redissolved in approximately 20 microliters of nonane. The 13 Clabeled recovery standards for PCDD/Fs (EPA-1613ISS, Wellington Laboratories Inc., Canada) were injected before instrumental analysis. Concentrations of PCDD/Fs were analyzed by high-resolution gas chromatography tandem high-resolution mass spectrometry (HRGC-HRMS, DFS, Thermo Fisher Scienti c, U.S.), equipped with a capillary column (DB-5 MS, 60 m × 0.25 mm ID × 0.25 µm, Agilent Technologies, U.S.). The total lipids of each maternal serum sample were detected by an enzymatic summation method (Covaci A, Voorspoels S, Thomsen C, van Bavel B, & Neels H, 2006). All data were reported on a lipid basis.

Quality control
One procedure blank test was carried out for every eight serum samples to examine contamination of the analysis system, and one test of standard reference material (Organic Contaminants in Non-Forti ed Human Serum, SRM-1957) purchased from the U.S. National Institute of Standards and Technology.
Procedure blank and SRM-1957 were analyzed to identify potential system contamination and as a quality control to assess laboratory precision during the entire analytical process. The precision and accuracy of measurements for PCDD/Fs in SRM-1957 were within the certi ed reference ranges recommended by the manufacturer. The method detection limits (MDLs) were 0.05-0.16 pg/g lipid. Recoveries of internal standards ranged from 47-99%, thus meeting the requirements of U.S.
We considered multiple covariates and potential confounders for the association of PCDD/F exposure with in uencing factors and BP, and factors with p < 0.1 were included in the nal multiple linear regression models. The following variables were considered for use: maternal age (years), body mass index (BMI, kg/m 2 ), days of pregnancy, education level (junior high school, high school, college and above), average monthly family income (< 5000 yuan, ≥ 5000 yuan), parity (0, ≥ 1), alcohol consumption (No, Yes), active smoking (No, Yes), passive smoking (No, Yes), occupation (worker, farmer, others), and nationality (Han nationality, others). Values were considered statistically signi cant when p ≤ 0.05.

Demographic characteristics of the study participants
The demographic characteristics of the study participants are shown in Table 1. The mean with standard deviations (SD) of maternal age, pregnancy BMI, and days of pregnancy were 27.98 ± 3.55 years old, 21.55 ± 3.29 kg/m 2 , and 84.12 ± 8.71 days, respectively. More than half of the mothers (54.8%) were having their rst pregnancy. The majority of the mothers (56.7%) had a high school education. Nearly one-third (36.4%) of the mothers did not work during pregnancy. Few mothers were active smokers (2.0%) or consumed alcohol (16.4%), but 43.3% of mothers passively smoked more than 15 min per day.

PCDD/Fs concentrations in maternal serum
The concentrations and detection rates of PCDD/Fs in maternal serum are shown in Table 2. The detection rates ranged from 32% (2,3,7,8-TCDD) to 100% (OCDD), ∑PCDD/F concentrations in maternal serum ranged from 163.52 pg/g lipid to 1,513,949.52 pg/g lipid, with a mean of 10,474.22 pg/g lipid, and OCDD (10,111.19 pg/g lipid) was the highest concentration. ∑PCDD concentrations (10,321.86 pg/g lipid) were signi cantly higher than ∑PCDFs (152.36 pg/g lipid). The mean ∑TEQ-PCDD/Fs was 42.3 pg/g lipid, and the ∑PCDDs were 2-fold higher than the ∑PCDFs.

Other factors
In uencing factors on PCDD/F concentrations and TEQs [β, 95% con dence interval (CI)] are presented in Table 3. Pregnant women with college and above education had higher concentrations of ∑PCDD/Fs than those with junior high school and below (β = 0.34, 95% CI: 0.01, 0.67). No associations were observed between the other compounds and in uencing factors (Supporting Information (SI) Table S1).

Discussion
We measured the PCDD, PCDF, PCDD/F, and TEQ concentrations in maternal serum early in pregnancy for a Chinese birth cohort study (H-YCCS). The total TEQs of PCDDs, PCDFs, and PCDD/Fs in maternal serum between this study and other studies globally are shown in Table 5. Mean ∑TEQ-PCDDs, ∑TEQ-PCDFs, and ∑TEQ-PCDD/Fs in the present study were higher than most background concentrations reported in the serum of pregnant women in other locations of the world (Baba et al., 2018;Bao et al., 2020;Boda et al., 2018;Bruce-Vanderpuije et al., 2019;Rodríguez Dozal S et al., 2011;Todaka et al., 2010;Yu DZ et al., 2019). ∑TEQ-PCDD/Fs in maternal serum were also compared with the biomonitoring equivalent (BE) for risk assessment. BE values are designed to be used as screening tools to assess whether chemicals have a large, small, or no margin of safety compared to existing health-based exposure guidelines . On the basis of neurodevelopmental effects, ∑TEQs of 15 pg/g lipid in serum samples is consistent with the minimal risk level recommended by the Agency for Toxic Substances and Disease Registry (Aylward LL, LaKind JS, & Hays SM, 2008). Referring to the ∑TEQ-PCDD/Fs in our data, 175 (57.4%) of the participants showed higher levels than the BEs. Thus, future studies should address long-term exposure to dioxins in pregnant women .
Concentrations of tetrachlorinated to octa-chlorinated congeners in serum enriched with increasing amounts of chlorine. These accumulations of highly chlorinated congeners in serum might be attributed to their high molecular weight (Mannetje et al., 2012;Suzuki, 2005;Wittsiepe et al., 2007). It has been reported that once higher chlorinated congeners enter serum, they are more likely to accumulate with time (Consonni D, Sindaco R, & Bertazzi PA, 2012). Both the number and position of chlorine substituted on the ring of PCDD/Fs was important for metabolism (Pluess N, Poiger H, Hohbach C, & Schlatter C, 1988).
We found that pregnant women with a higher education had higher concentrations of ∑PCDD/Fs. Educational status may be a marker for socioeconomic status in this population, and both education and socioeconomic status have been positively associated with maternal dietary quality (Rifas Shiman SL, Rich Edwards JW, Kleinman KP, Oken E, & Gillman MW, 2009). They have better conditions and tend to increase the consumption of animal-origin food (Chen et al., 2013). It has been estimated that more than 90% of current human exposure to dioxins among the general population occurs via food consumption, primarily from animal-origin food (Ucar et al., 2011). PCDD/Fs are highly lipophilic, environmentally persistent, and more likely to be enriched in adipose tissues (Baars AJ et al., 2004).
Only a few studies have reported the relationship between PCDD/Fs and BP. Vietnam soldiers exposed to high levels of 2,3,7,8-TCDD may have an increased risk of hypertension but not other PCDD/F congeners (Cypel YS et al., 2016). It has been reported that high TEQs of PCDD/Fs in Japanese general populations may increase the risk of hypertension (Nakamoto et al., 2013). However, the relationship between dioxin exposure and PP during pregnancy has not been reported until now. In this study, high exposure to PCDD/Fs was associated with decreased PP in pregnant women. Our results differ from the conclusions reported by Cypel and Nakamoto. Animal studies have shown that dioxins could affect BP. Sustained aryl hydrocarbon receptor (AhR) activation by 2,3,7,8-TCDD exposure induces hypertension in adult male C57BL/6 mice. This model provides valuable insight into the mechanisms underlying 2,3,7,8-TCDD-induced cardiovascular pathogenesis, including the role of vascular reactive oxygen species (ROS) as potential mediators of 2,3,7,8-TCDDMinduced hypertension (Kopf PG et al., 2008). When C57BL/6J mice are continuously exposed to 2,3,7,8-TCDD, AHR regulates the expression of several genes, including CYP1A1 and cyclooxygenase (COX-2), which may in uence BP through the production of vasoactive eicosanoids (Timothy P. Dalton et al., 2001). However, different results showed that BP in the infected group was lower than that in the control group after exposure to 2,3,7,8-TCDD in female rats (Hermansky SJ et al., 1988). These studies suggest that 2,3,7,8-TCDD exposure could affect the BP of experimental animals, but the effect may be related to the dose or the species of the tested animals.
Mechanisms that might be involved in the associations between PCDD/Fs and PP are currently largely unknown. Experimental evidence suggests that dioxin-like compounds can modulate endothelium-derived vasoactive factors in human primary endothelial cells and induce changes characteristic of endothelial dysfunction in human essential hypertension (Valera B, Ayotte P, Poirier P, & Dewailly E, 2013). It can modulate the expression of vasoconstriction factors such as COX-2, prostaglandins, and reactive oxygen species (ROS) and change the production of nitric oxide (NO), a well-known vasodilator factor (Andersson H, Garscha U, & Brittebo E, 2011).
It has been reported that a decreased PP may be associated with hypovolemia, cardiac failure, cardiac arrhythmia, valvular heart disease, aortic dissection, or low BP (Muller M, 2010). Upon review of previous medical literature, a low PP has been shown to be an indicator of decreased cardiac function and poor outcomes in patients with myocardiac infarction and a predictor of cardiovascular death in patients with mild to advanced heart failure (El-Menyar A, 2011).
Although low PP was less apparent in pregnant women, if these associations are indeed present at relatively common serum PCDD/Fs concentrations, the public health signi cance of the relationships may be considerable. Clearly, it is very important that these possibilities should be tested systematically in a prospective investigation. There is also a need for investigations that can identify the mechanisms that might underlie these associations. For PP, a number of external factors can in uence levels or performance, but the observation that serum PCDD/Fs levels can in uence them indicates that PCDD/Fs levels may cause physiological changes and contribute to the development of diseases even when they are not the sole cause.
This study has several limitations. First and foremost is the limited sample size of our study population.
As the serum dioxin concentration is at the pg/g level, su cient serum is required. Only a subset of the population provided su cient samples. Second, the study on the relationship between dioxin exposure and PP is a cross-sectional study, which cannot prove causality regardless of how strong the associations are.

Conclusions
Our collective ndings reveal that pregnant women may be at risk of exposure to PCDD/Fs. The ratio of tetrachlorinated to octa-chlorinated congeners in maternal serum was enriched with an increasing number of chlorines. Pregnant women with higher education had higher levels of ∑PCDD/Fs. We found that ∑TEQs were signi cantly and negatively associated with PP. These exploratory results require corroboration with further studies involving larger sample sizes and a wider range of exposure to establish the association between exposure to dioxins and maternal as well as offspring health.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. renamedf2fbc.docx