Arch Toxicol DOI 10.1007/s00204-015-1515-8

INORGANIC COMPOUNDS

Revisiting mobilisation of skeletal lead during pregnancy based on monthly sampling and cord/maternal blood lead relationships confirm placental transfer of lead Brian Gulson1,2   · Karen Mizon1 · Michael Korsch2 · Alan Taylor3 

Received: 13 January 2015 / Accepted: 31 March 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  Lead (Pb) can be released from the maternal skeleton during pregnancy and lactation and transferred to the infant. Most support for this hypothesis comes from blood Pb (PbB) studies involving limited sampling during pregnancy, the maximum usually being five samplings, including at delivery. We provide longitudinal data for PbB concentrations and Pb isotopic ratios for three cohorts of pregnant females (n  = 31), two of which are based on monthly sampling and the other on quarterly sampling. We also provide data for samples collected post-partum. The data are compared with changes observed in a matched, by country and age, non-pregnant control cohort (n  = 5). The monthly data illustrate the variability between subjects, which is also apparent when the data are compared on a trimester basis. Mixed model analyses showed that, This paper is dedicated to the memory of three wonderful people, Dr. Mahaffey, who was our first US National Institute for Environmental Health Sciences project officer for several years, Professor Tony McMichael, who led the first stage of the contract, and Mary Salter, who was our phlebotomist for 13 years. Electronic supplementary material  The online version of this article (doi:10.1007/s00204-015-1515-8) contains supplementary material, which is available to authorized users. * Brian Gulson [email protected]

in the third trimester, the mean PbB level was significantly lower for women (n = 10) who took a calcium (Ca) supplement (PbB 1.6 µg/dL) than those whose Ca intake was low (low-Ca cohort; n  = 15; PbB 2.5 µg/dL) because low Ca means more mobilisation is required for homoeostasis so that more Pb was mobilised from the skeleton. For women who took the supplement, post-partum PbB levels were significantly higher than those in the other periods (2.7 vs 1.4–1.6 µg/dL). For women in the low-Ca cohort, PbB levels were higher at post-partum than in pre-pregnancy and in the first and second trimesters (3.1 vs 1.8 µg/dL), while the levels in the third trimester were higher than those in the first and second trimesters. Importantly, the increase in PbB during gestation was delayed until the third trimester in the Ca-supplemented cohort compared with the low-Ca cohort. Regression analysis showed that the changes over trimester were very similar for PbB and the 206Pb/204Pb ratio providing convincing evidence for extra mobilisation of Pb from the maternal skeleton during pregnancy and lactation. Isotopic ratios in the cord blood samples were similar to those in the maternal blood samples taken prior to parturition with an R2 0.94 for the migrant subjects and R2 0.74 for Australian subjects for 206Pb/204Pb ratios, supporting the concept of placental transfer of mobilised skeletal stores of Pb. Keywords  Lead · Blood · Skeletal lead · Pregnancy · Post-partum · Isotopic ratios · Cord blood · Calcium supplementation

1

Faculty of Science and Engineering, Graduate School of the Environment, Macquarie University, Sydney, NSW 2109, Australia

2

Energy Flagship, Commonwealth Scientific and Industrial Research Organisation (CSIRO), Sydney, NSW 1670, Australia

Introduction

3

Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

The toxic metal lead (Pb) is mainly stored in the bone, amounting to more than 90 % in adults (Barry 1975) but







13



can be released during times of physiological stress such as pregnancy, lactation, and menopause (Silbergeld 1991), extended bed rest and weightlessness. The most detrimental effects of Pb are on the developing infant during pregnancy and early childhood (US EPA 2013). It has been shown that Pb released from the maternal skeleton during pregnancy and lactation can be transferred to the infant with the strongest evidence coming from investigations using stable isotopes of Pb (Franklin et al. 1997; Gulson et al. 1997a, 1998a, b, 2003, 2004; Manton 1992; Manton et al. 2000, 2003). Indirect evidence for mobilisation of Pb during pregnancy and lactation has come from changes in blood Pb (PbB) concentrations (Ettinger et al. 2009; Farias et al. 1996; Hernandez-Avila et al. 2003; Hertz-Picciotto et al. 2000; Lagerkvist et al. 1996; Liu et al. 2013; Moura and Goncalves Valente 2002; Rothenberg et al. 1994, 1999; Schell et al. 2000, 2003; Sowers et al. 2002; Tellez-Rojo et al. 2004; West et al. 1994), Schumacher et al. 1996). Changes in PbB during pregnancy in these studies were usually based on limited sampling, commonly restricted to a single sample for three trimesters or for the second and third trimesters or the first and third trimesters. In some cases, the results were not from the same subjects during the course of pregnancy. Sampling of up to five times during pregnancy was undertaken by Hertz-Picciotto et al. (2000) and four times by Schumacher et al. (1996) and Sowers et al. (2002). Where samples were measured for the same subjects over three trimesters, the PbB concentrations showed a U-shaped curve with the minimum value at about 20 weeks (Hertz-Picciotto et al. 2000; Rothenberg et al. 1994). Rather than relying solely on PbB concentrations, which can vary, not only because of mobilisation of Pb from bones during pregnancy, but also because of changes in haematocrit values, combining the PbB values with Pb isotopic ratios adds another dimension of certainty to interpretations as the Pb isotopic ratios are unaffected by changes in haematocrit. Another reason changes in PbB are not reliable indicators of mobilisation of bone stores is that pregnancy and lactation alter other aspects of Ca metabolism, such as gastrointestinal uptake and renal clearance. We have published overall syntheses of our Pb investigations during pregnancy and post-partum (see some of these in the first paragraph) but not details for individuals nor rigorous statistical evaluation analyses of the changes in PbB and PbB-Pb isotopic relationships. We have obtained longitudinal Pb data on a monthly basis during pregnancy for migrant subjects and on a trimester basis for Australian subjects, in addition to post-natal data. In this paper, we provide the detailed results for 35 individuals from three sets of pregnant female subjects (2 migrant cohorts to Australia and a control cohort of long-term Australian subjects)

13

Arch Toxicol

and five non-pregnant migrant control subjects. We compare the outcomes obtained for PbB concentrations and Pb isotopic ratios, which illustrate the strength of using isotopic measurements. In addition, we provide a comparison of maternal and cord blood Pb isotopic and concentration data, which confirms placental transfer of Pb.

Methods Lead in the environment is dominated by industrial sources, which in turn have been derived from relatively few mineral deposits, most of which have distinctive isotopic patterns (signatures or fingerprints). People growing up in a particular environment will take Pb into their bodies with a characteristic signature. A proportion of this Pb will be incorporated into their skeletons that are then effectively labelled with this signature. If an individual moves to an environment with a different Pb signature, the original signature can be detected in the individual’s blood for many years because Pb is released from the skeleton as part of the normal bone remodelling. What is measured in the blood (and urine) is a mixture of the signatures from the new and old environment. Any process that affects bone turnover rate will affect the proportions of the two signatures. The prevalence of geologically old Pb (~1700 million years old) in the Australian environment has resulted in a unique isotopic fingerprint or signature, generally not available on other continents. We have established from blood analyses of almost 300 subjects that the Pb isotopic fingerprint in subjects from other countries is significantly different from those of multi-generational Australian residents. By monitoring the PbB isotopes of migrant subjects after arrival in Australia, it is possible to detect changes in isotopic composition and PbB concentration related to mobilisation of skeletal Pb during pregnancy and lactation (see references above) and ageing (Gulson et al. 2002). Hence, we studied female subjects from mostly European countries, whose Pb isotopic signature in their skeleton and their blood was imprinted with Pb exposure from different sources to that prevailing in the Australian environment. Details of subjects, sampling methods, and analyses have been described in our earlier publications (Gulson et al. 2003). Subjects (Table 1) Four cohorts were investigated, three in which subjects conceived. One cohort, termed the low-Ca cohort, comprised migrant subjects (n  = 15) whose daily intakes of Ca were less than half the recommended guidelines of 1000 mg. The second cohort, termed the Ca-supplemented cohort, comprised migrant subjects (n  = 10) who were

Arch Toxicol Table 1  Subject information Identifier

Birth country

208

Low-Ca cohort 1009 1016 1022 1035 1041 1042 1043 1045 1052 1055 1056 1069 1090 1096 1097

Croatia Armenia Ukraine Poland Bosnia Bulgaria Poland Poland Poland France Bosnia Bosnia China China Romania

Ca-supplemented cohort 1204 1208

Pb/206Pb

207

Pb/206Pb

206

Pb/204Pb

Pb B (µg/dL)

First date collected

2.1268 2.1145 2.1507 2.1135 2.1187 2.0832 2.1049 2.0945 2.0981 2.1086 2.1123 2.1112 2.1250 2.1183 2.0995

0.8835 0.8707 0.8806 0.8724 0.8774 0.8439 0.8669 0.8571 0.8602 0.8661 0.8676 0.8663 0.8727 0.8689 0.8599

17.58 17.91 17.63 17.89 17.73 18.51 17.99 18.22 18.16 18.03 18.01 18.03 17.86 17.96 18.14

3.05 2.25 2.14 1.47 1.94 20.00 1.96 2.70 1.92 2.55 3.11 3.13 2.55 2.15 3.22

05-Nov-93 16-Feb-94 02-Mar-94 16-Jun-94 14-Sep-94 29-Jun-92 31-Oct-94 25-Nov-94 31-Mar-95 15-Jun-95 15-Jun-95 09-Nov-95 26-Apr-96 13-Dec-96 10-Dec-96

Bulgaria Bosnia

2.0984 2.1134

0.8575 0.8739

18.19 17.81

3.98 1.91

01-Dec-98 28-Jul-99

1211 1212 1213 1214 1225 1226 1229 1231

Bangladesh Turkey Lebanon Turkey Pakistan Iraq Lebanon China

2.1269 2.1298 2.1222 2.1088 2.1270 2.1048 2.1000 2.1145

0.8832 0.8836 0.8825 0.8702 0.8826 0.8681 0.8684 0.8641

17.60 17.62 17.61 17.93 17.63 17.98 17.95 18.06

1.58 2.38 2.89 1.83 6.49 1.43 2.34 1.72

16-Aug-99 20-Aug-99 01-Sep-99 22-Sep-99 09-Apr-01 07-May-01 22-May-01 29-May-01

Australian cohort 1049 1057 1065 1066 1085 1093

Australia Australia Australia Australia Australia Australia

2.1692 2.1533 2.1715 2.1571 2.1600 2.1612

0.9187 0.9030 0.9191 0.9092 0.9088 0.9106

16.84 17.20 16.88 17.03 17.08 17.03

4.28 3.69 2.99 2.56 1.85 4.00

17-Feb-95 07-Jun-95 13-Sep-95 02-Nov-95 21-Nov-95 19-Jun-96

Non-pregnant migrant cohort 1015 1031 1047 1064

Russia Bosnia Ukraine Russia

2.1043 2.1075 2.1214 2.1393

0.8648 0.8659 0.8767 0.8923

18.02 18.01 17.75 17.40

3.42 3.52 2.30 2.47

16-Feb-94 27-May-94 15-Feb-95 07-Sep-95

1224

Bulgaria

2.0961

0.8575

18.18

2.43

24-Jan-01

PbB first sample taken on recruitment

administered about 1200 mg Ca daily from supplements. The third cohort comprised long-term Australian subjects (n  = 6), two of whom conceived twice during the study. A fourth cohort comprised migrant subjects who were matched with a non-pregnant subject on the basis of country of origin, age, and number of pregnancies (n = 5).

Sampling Monthly blood samples by venipuncture were obtained from migrant and Australian subjects during pregnancy and the post-partum period for 6+ months. At parturition, cord blood was obtained from all subjects. Environmental

13



Arch Toxicol 18.0

samples (6-day duplicate diet, drinking water, house dust, gasoline, ambient air) were collected quarterly. These data have been published in several papers (Gulson et al. 1997b, 1999, 2001, 2006).

1

PbB

17.0

0

200

400

600

800

13

0 1000

Days after Arrival in Australia

Fig. 1  Changes in 206Pb/204Pb and PbB for migrant subject 1022 from the low-Ca cohort. The cord blood data are indicated by crosses

1045

5 4

Pb/204Pb

206Pb/204Pb

18.0

PbB

3 2

17.5

PbB (µg/dL)

18.5

1 17.0

0

200

400

600

800

0 1000

Days after Arrival in Australia

Fig. 2  Changes in 206Pb/204Pb and PbB for migrant subject 1045 from the low-Ca cohort. The cord blood data are indicated by crosses

1056

18.0

5 4 3

17.5 2 206Pb/204Pb

1

PbB

17.0

0

200

400

600

800

PbB (µg/dL)

The number of samples measured for migrant subjects during gestation ranged from 5 to 10, including the cord blood sample. For the Australian subjects, only trimester and cord blood samples were usually measured as initial analyses showed the gestational data to be relatively uniform. To illustrate the detailed changes over time, results for PbB and the Pb isotopic ratio 206Pb/204Pb are presented in Figs. 1, 2, 3, 4, 5, 6, 7, 8, and 9 for three subjects from each pregnant cohort. Because of space limitations, the results for other subjects are given in the Supplementary Notes (Figures S1-S20). Results for the other two isotopic ratios (208Pb/206Pb and 207Pb/206Pb) are similar to the 206Pb/204Pb ratios, and, if interested to a reader, the data for the other

PbB (µg/dL)

Pb/ 204 Pb 206

2

206

Results

3 17.5

206Pb/204Pb

206

The effect of the Ca supplement on PbB in the migrant women during pregnancy and post-partum was investigated using a mixed model (SPSS v22) with PbB as the dependent variable and trimester (before, during trimesters 1, 2, and 3, and post-partum), which was treated as a categorical variable with five levels, calcium supplement (0 for no supplement and 1 for supplement), and their interaction, as the independent variables. Time, measured in months, a numeric variable which had a value of zero for the first observation for each subject, and Food Pb, the level of Pb in the women’s diets, were included as covariates. Subject was treated as a random factor to take account of the correlations between the multiple observations for each subject. The relationship between isotopic ratio expressed as the 206 Pb/204Pb ratio and PbB over time for the low-Ca pregnant cohort was investigated using a regression analysis with trimester (excluding zero) and type of measurement as the independent variables and lead level as the dependent variable. Both types of measurements (PbB and the ratio) were standardised so that, within each subject, each type of measurement had a mean of zero and standard deviation of one. This meant that all between-subject variation was removed (the mean of the four observations for each subject was zero); this facilitated comparisons of changes over time. The primary interest was in the main effect of trimester and the interaction of trimester and type of measurement. As the measurements for each subject were aggregated (separately for each type of measurement and trimester), there was no need for a random factor.

4

Pb/204Pb

Statistical analyses

5

1022

1000

1200

0 1400

Days after Arrival in Australia

Fig. 3  Changes in 206Pb/204Pb and PbB for migrant subject 1056 from the low-Ca cohort. The cord blood data are indicated by crosses

isotopic ratios can be obtained from the senior author. Precision of the 206Pb/204Pb ratio is ±0.2 % (2 sigma) and for PbB concentration is ±3 % (2 sigma) for a blood sample containing

maternal blood lead relationships confirm placental transfer of lead.

Lead (Pb) can be released from the maternal skeleton during pregnancy and lactation and transferred to the infant. Most support for this hypothesis co...
622KB Sizes 2 Downloads 5 Views