Distribution of Poly-and Perfluoroalkyl Substances in Matched

Jun 18, 2013 - College of Environmental Sciences and Engineering, Ministry of ... and Center for Environment and Health, Peking University, Beijing 10...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/est

Distribution of Poly- and Perfluoroalkyl Substances in Matched Samples from Pregnant Women and Carbon Chain Length Related Maternal Transfer Tao Zhang,† Hongwen Sun,*,† Yan Lin,‡ Xiaolei Qin,† Yanfeng Zhang,† Xia Geng,§ and Kurunthachalam Kannan∥ †

College of Environmental Sciences and Engineering, Ministry of Education Key Laboratory of Pollution Processes and Environmental Criteria, Tianjin Key Laboratory of Environmental Remediation and Pollution Control, Nankai University, Tianjin 300071, China ‡ College of Environmental Sciences and Engineering and Center for Environment and Health, Peking University, Beijing 100871, China § Waters Technologies (Shanghai) Ltd., Shanghai 201203, China ∥ Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Albany, New York 12201, United States S Supporting Information *

ABSTRACT: Although levels of poly- and perfluoroalkyl substances (PFASs) in human maternal and neonatal blood have been widely reported in the literature, relationship of maternal−fetal transmission of PFASs with carbon chain length is presently not well understood. In this study, 11 PFASs were analyzed in matched samples, including not only maternal blood (MB, n = 31) and cord blood (CB, n = 30), but also placenta (n = 29) and amniotic fluid (AF, n = 29). Except for perfluorohexanoic acid (PFHxA), the detection frequencies of PFASs were similar among placenta, MB, and CB (>80% for 8 PFASs, nondetectable for 2 PFASs). Though only perfluorooctanoic acid (PFOA) was frequently detected (>90%) in AF, with a median concentration of 0.043 ng/mL, other 5 PFASs were also detectable in AF samples with low concentrations (mean: 0.013−0.191 ng/mL). This suggests that in addition to blood-borne in utero exposure, the fetus is also exposed to low levels of PFASs through AF. Concentrations of PFOA in AF were positively correlated with those in MB (r = 0.738, p < 0.01) and CB (r = 0.683, p < 0.001), suggesting that AF concentration could reflect fetal PFOA exposure during pregnancy and can be used as a biomarker. To clarify the effects of carbon chain length on maternal transfer of PFASs, we calculated maternal transfer efficiencies of PFASs from MB to CB (TMB‑CB). A U-shaped trend in TMB‑CB of C7−C12 perfluoroalkyl carboxylic acids (PFCAs) with increasing carbon chain length was found in this study for the first time. The U-shaped TMB‑CB of PFCAs with carbon chain length is an integrated result of opposite trend of the ratios between MB/placenta and placenta/CB based on carbon chain length. This is the first study to report the occurrence of PFASs in human placenta. The results reported here enable better understanding of the maternal−fetal transmission of PFASs.



INTRODUCTION

and development, even at low levels of PFOS (range: 1.3−16.2 ng/mL in maternal serum);7−9 while these associations were not found in other studies.10,11 Available evidence suggest that toxicity and bioaccumulation of PFASs increase with increasing carbon chain length,12−14 whereas short-chain PFASs show a higher mobility in the environment.15,16 Therefore, characterization of the maternal−

Exposure to toxic chemicals during critical periods of development, including the prenatal period, could potentially influence growth, reproduction, and development of humans.1,2 Poly- and perfluoroalkyl substances (PFASs) are a family of contaminants that have drawn much attention recently due to their extensive detection in environmental and human samples.3−5 Perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) are two PFASs of most concern, and toxicity studies in animals have indicated the potential for PFOA and PFOS to affect maternal fecundity.6 Several studies have reported negative associations of PFASs with fetal growth © 2013 American Chemical Society

Received: Revised: Accepted: Published: 7974

March 1, 2013 June 13, 2013 June 18, 2013 June 18, 2013 dx.doi.org/10.1021/es400937y | Environ. Sci. Technol. 2013, 47, 7974−7981

Environmental Science & Technology

Article

The maternal age ranged from 21 to 39 years (mean: 30.9 years); gestational age ranged from 35 to 47 weeks (mean: 40.3 weeks). Before delivery, the subjects’ body weight was between 59 and 105 kg, with a mean value of 74.7 kg. All subjects had full-term pregnancies. Overall, 82% of women were primiparous, and 18% gave birth to their second child. All participants were healthy, and none of the participants reported occupational exposure to PFASs. Detailed demographic information of the subjects is shown in Table S1. During May to June 2010, we collected 29 placenta, 29 AF, 31 MB, and 30 CB with 27 matched samples from the same subject/fetus (Figure S1). Whole blood was collected in this study. Placenta, AF, and CB were collected at the time of delivery. MB samples were collected in the preoperative holding area within one hour of delivery. Placenta was stored in a hermetic polyethylene bag; AF samples were collected in 50-mL polypropylene tubes; blood samples were collected in heparinized vacutainers. In the laboratory, placenta was freezedried immediately and stored at −20 °C; AF and blood samples were frozen at −20 °C until analysis. The study was approved by the Institutional Review Board of Nankai University, China. Sample Preparation and Analysis. All placenta and blood samples were extracted by ion-pair extraction procedure as described earlier;32 AF samples were extracted using Oasis WAX SPE cartridge (Waters Corp, Milford, MA, USA), and were purified using Envi-carbon cartridge (Supelco, Inc., Bellefonte, PA, USA). Concentrations of 11 PFASs were determined with Waters Acquity ultra-performance liquid chromatography equipped with Waters Acquity TQD triple quadrupole mass spectrometer (UPLC-MS/MS). Details regarding reagents and chemicals, sample preparation, and instrumental analysis are given in the SI. Quality Assurance and Quality Control. Matrix-spike recoveries of individual PFASs through the analytical procedure were calculated by spiking 11 target compounds into randomly selected placenta and blood samples (10 ng each). Recoveries of the PFASs spiked into sample matrices ranged from 70 ± 8% to 120 ± 8% for placenta (n = 3) and from 84 ± 12% to 117 ± 7% for whole blood (n = 5) (Table S2). A few exceptions, with low/high recoveries were noted: perfluorobutane sulfonate (PFBS) in placenta and whole blood, perfluorohexanoic acid (PFHxA), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluorododecanoic acid (PFDoDA) in placenta, and PFDA and PFDoDA in whole blood showed abnormal recoveries either below 50% or greater than 130%. Therefore, caution needs to be exercised when discussing the concentrations of these PFASs in corresponding sample matrices. As internal standards, sodium perfluoro-1[1,2,3,4-13C8]octanesulfonate (MPFOS) and perfluoro-n[13C8]octanoic acid (M8PFOA) were spiked (5.0 ng each) into all samples prior to extraction. Respective mean recoveries of MPFOS and M8PFOA were 81 ± 7% and 78 ± 6% for placenta and 102 ± 7% and 98 ± 9% for whole blood (Table S2). Method precision was good, with relative standard deviations (RSDs) for each sample matrix in the range of 1− 17% for all PFASs (Table S2). Matrix spikes were prepared by spiking known amounts of PFASs (3.0 ng each) into randomly selected AF samples. The recoveries of PFOS and PFOA were 55 ± 8% and 57 ± 9% (n = 3), respectively (Table S2). Although the recoveries of PFOS and PFOA were low for AF, corresponding internal standards (i.e., MPFOS and M8PFOA) were spiked (level: 0.10 ng/mL in AF; n = 29) into all samples before extraction. The mean

fetal transmission of PFASs with varying carbon chain length is critical for the assessment of risks and mechanisms in developmental toxicity. A German study on PFASs in matched maternal−fetal blood suggested that the maternal transfer efficiency of long-chain PFASs was lower than that of short chain PFASs.17 Similarly, a decreasing trend in maternal transfer efficiency of PFASs with carbon chain length was also reported from Korea, Canada, and South Africa.18−20 However, such carbon chain length related decrease in the maternal transfer was not found in Norway, and in another Korean study.9,21 Although several studies have reported PFAS concentrations in maternal−fetal blood, there is still no clear understanding of the maternal transfer of PFASs with different carbon-chain length.9,17−24 The partitioning ratios between maternal and fetal tissues of different chemicals varied greatly and may depend on physical− chemical properties. It has been found that PFASs more readily cross the placenta membrane to the developing fetus compared to BPA;25,26 while PFASs are more efficiently blocked by the placental barrier than are PBDEs and heavy metals.27,28 Limited studies have shown that PFASs, unlike traditional persistent organic pollutants (most of them are lipophilic chemicals), are proteinophilic.29,30 The placenta supplies the fetus with oxygen and nutrients; however, many toxic chemicals including PFASs, can be transferred via placenta to the fetal circulation. Therefore, placenta can be used as an indicator of chemicals’ exposure for both mother and fetus.31 Previous studies measured PFASs in matched maternal blood (MB), cord blood (CB), breast milk, and infant blood samples.7−11,17−24 The partition ratios of PFASs between MB and CB as well as MB and breast milk have been well documented. However, there is a general lack of information on placental levels of PFASs, and distribution of PFASs between MB (or CB) and placenta. To our knowledge, this is the first study to report PFAS levels in human placenta. Moreover, amniotic fluid (AF) surrounds and protects the developing embryo and fetus, and the composition varies over the course of gestation. Although the fetal lungs and kidneys excrete fluid continuously into the AF reservoir, the fetus is also continuously ingesting and inhaling the AF. This cycling suggests that chemicals excreted into the AF can be re-exposed to the fetus. Therefore, measurements of environmental chemicals in AF may be a useful biomarker of fetal exposure. In this study, we analyzed 11 PFASs in matched placenta, AF, MB, and CB samples collected in Tianjin, China. The biomonitoring of PFASs in placenta further clarified maternal transfer of PFASs; PFAS levels in CB and AF reflect fetal exposure during the course of pregnancy. The principal objectives of this study were to (1) report PFAS residue levels in matched placenta, AF, MB, and CB; (2) assess the associations of PFAS concentrations among placenta, AF, MB, and CB; and (3) elucidate the effects with different carbon chain lengths on maternal transfer efficiencies of perfluoroalkyl carboxylic acids (PFCAs).



MATERIAL AND METHODS Study Participants and Sample Preparation. Thirty two pregnant women and their fetuses were recruited at a hospital in Tianjin, China. Subjects were identified before admission to hospitals, and the subjects’ physicians were asked for permission to offer participation. We obtained characteristics of the subjects from self-administered questionnaire surveys and medical records (Supporting Information (SI) Table S1). 7975

dx.doi.org/10.1021/es400937y | Environ. Sci. Technol. 2013, 47, 7974−7981

Environmental Science & Technology

Article

Table 1. Concentrations of Poly- and Perfluoroalkyl Substancesa in Matched Maternal Blood, Placenta, Cord Blood, and Amniotic Fluid Samples from China (ng/mL for Blood and Amniotic Fluid, ng/g fw for Placenta) c

chain length

PFHxS

PFOS

PFPeA

PFHxA

PFHpA

PFOA

PFNA

PFDA

PFUnDA

PFDoDA

ΣPFASsb

C6

C8

C5

C6

C7

C8

C9

C10

C11

C12

NAe

100 7.39 36.1 12.4 14.6 4.98

0