Kinetic differences of legacy organochlorine pesticides and polychlorinated biphenyls in vietnamese human breast milk

1 Nguyen Minh Tue a, Agus Sudaryanto b, Tu Binh Minh c, Bui Hong Nhat c,Tomohiko Isobe b, Shin Takahashi a, Pham Hung Viet c, Shinsuke Tanabe a,* a Matsuyama 790+8577, Japan b Center for Marine Environmental Studies, Ehime University, Bunkyo+cho 2+5, Senior Research Fellow Center, Ehime University, Bunkyo+cho 2+5, Matsuyama 790+ 10 8577, Japan 11 c 12 of Science, 334 Nguyen Trai, Hanoi, Vietnam Centre for Environmental Technology and Sustainable Development, Hanoi University 13 14 * 15 University, 2+5 Bunkyo+cho, Matsuyama 790+8577, Japan Tel./Fax: +81 89 927 8171 16 E+mail: shinsuke@agr.ehime+u.ac.jp (S Tanabe) Corresponding author Center for Marine Environmental Studies (CMES), Ehime 17 18 19 The present study investigated the current contamination status and evaluated several 20 kinetic+related features of organochlorine pesticides (OCPs) and PCBs in human breast 21 milk collected from northern Vietnam The variation in the levels of these contaminants 22 was found to be strongly associated with total lactation time and dietary habits OCPs 23 exhibited the characteristics of steadily declining compounds: the overall levels of 24 DDTs and HCHs in the population decreased with a half+time of only years and it can 25 be suggested that OCPs depurated relatively fast with breastfeeding (5% per month) 26 PCBs were slower in both regards, with a temporal decrease half+time of 12 years and a 27 suggested depuration rate via breastfeeding of 2.5% per month, indicating that the 28 exposure level was still high relative to the human body burden It was found that the 29 PCB exposure levels of infant from breastfeeding exceeded the reference dose, and this 30 situation may continue for the next two or three decades Knowledge of these kinetic+ 31 related characteristics not only is useful for risk assessment and prediction of future 32 trends of legacy contaminants but also may provide insight regarding similar kinetic 33 processes of emerging persistent pollutants 34 35 Keywords: breast milk, depuration, POPs, temporal trend, Vietnam 36 37 38 39 Worldwide production and intensive use of organochlorine compounds (OCs), 40 including organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs), has 41 resulted in their widespread contamination These compounds have received 42 considerable attention in the late decades of the last century with regard to their 43 persistence, bioaccumulative characteristics, long+range transport and long+term toxic 44 effects to human (ATSDR, 2000, 2002) Despite the global declining trends of OCs in 45 recent years (Kunisue et al., 2006; Jaraczewska et al., 2006; Lignell et al., 2009; Norén 46 and Meironyté, 2000; Schecter et al., 2005), elevated levels of OCPs were still observed 47 in human breast milk from Asian developing countries due to a later phase+out than in 48 developed countries, suggesting further monitoring is necessary (Kunisue et al., 2004; 49 Minh et al., 2004; Subramanian et al., 2007; Sudaryanto et al., 2006) Vietnam was 50 reported to be among the countries with the highest levels of 51 dichlorodiphenyltrichloroethane and its metabolites (DDTs) and unlike other Asian 52 developing countries also had relatively high levels of PCBs (Minh et al., 2004) because 53 of the extensive use of OCPs in agriculture and imported PCB+contaminated 54 transformer oil before their ban in 1995 (Minh et al., 2008) 55 Breast milk is a convenient matrix for monitoring persistent organic pollutants 56 (POPs) in human The advantages include simple and non+invasive sample collection, 57 suitability for determination of lipophilic POPs due to the relatively rich lipid content, 58 and relevance with regard to the exposure of breastfeeding infants, who are at the early 59 stage of development and vulnerable to toxic contaminants However, due to the 60 transfer of the mother's body burden to the infant through breastfeeding (Carrizo et al., 61 2007; Shen et al., 2007; Trapp et al., 2008), the contamination levels in female donors 62 may vary with lactation time Indeed, the levels of legacy POPs such as OCs in mother's 63 milk have been reported to decrease with the number of children (Fürst, 2006; Kunisue 64 et al., 2006; Minh et al., 2004; Sudaryanto et al., 2006) Recent monitoring studies have 65 been designed to avoid the variation by sampling milk from only primiparae within 3–8 66 weeks post+partum (WHO, 2005) With this approach, it is difficult however to make a 67 direct comparison with data from past references and the survey is also more difficult to 68 coordinate, especially in developing countries where health assessment frameworks are 69 scarce There have been studies aiming to quantify the variation due to lactation, either 70 by temporal monitoring of the compounds in individual donors (Hooper et al., 2007; 71 Schecter et al., 1998) or by using mathematical models (Trapp et al., 2008) Together, 72 these studies indicate that the observed depuration kinetic of POPs during the lactation 73 period varies for different compounds and suggest an important influence of human 74 intake level This influence may explain the lack of significant difference in the breast 75 milk levels of emerging POPs such as polybrominated diphenyl ethers (PBDEs) 76 between primiparae and multiparae mothers (Schecter et al., 2003; Tue et al., 2010) 77 Consequently, while the depuration kinetics of legacy and emerging POPs through 78 breastfeeding may be similar, investigating the former has the advantage because their 79 human intake levels are low relative to the body burden 80 The present study investigated the current contamination status of legacy OCPs and 81 PCBs in northern Vietnamese human breast milk and assessed their different kinetic+ 82 related characteristics, specifically the relationship between OC levels and lactation 83 time as well as the connection of this relationship to the declining trend of the 84 compound An up+to+date assessment of infant health risk from OCs in mother's milk 85 was also conducted 86 87 88 89 Breast milk samples ( = 33) were collected between August and September 2007 90 from Hanoi, the capital city of Vietnam and its surrounding suburban and rural areas 91 All donors were non+smokers, non+drinkers and appeared healthy The samples were 92 collected by the donors or with the help of a midwife and placed in solvent+pre+cleaned 93 analytical+grade glass containers with Teflon+lined screw caps and kept in gel ice 94 Informed consents were obtained from all donors Questionnaires were also used to 95 acquire information on age, height and weight measurements, number of deliveries, 96 detailed history of breastfeeding, occupation and dietary habit General information on 97 the donors are summarised in Table After collection the samples were kept with ice 98 and sent within h to the Centre for Environmental Technology and Sustainable 99 Development (Hanoi University of Science, Hanoi, Vietnam) to store at –20 °C The 100 frozen samples were later air+transported with gel ice to the Environmental Specimen 101 Bank (es+BANK, Ehime University, Japan) and stored at –25 °C until analysis 102 103 104 Analysis of OCs followed the method described by Minh et al (2004) Briefly, 105 approximately 10 g of sample was applied to a glass column containing 10 g of pre+ 106 cleaned diatomaceous earth (EXtrelut® NT, Merck, Germany), kept for 30 and then 107 extracted with 250 ml diethyl ether The extract was dried over anhydrous sodium 108 sulphate then concentrated and solvent+exchanged into hexane A portion of this extract 109 (25% in volume) was used for gravimetric determination of lipid content The 110 remaining extract was subjected to gel permeation chromatography (packed Bio+Bead 111 S+X 3, Bio+Rad Laboratories, USA) for lipid removal using a dichloromethane 112 (DCM)/hexane mixture (1:1 v/v) as eluant The lipid+free extract was then concentrated 113 and passed through a 12 g of activated Florisil (Florisil PR, Wako, USA) packed in a 114 glass column for final clean+up and separation The first fraction, eluted with hexane, 115 contained PCBs, hexachlorobenzene (HCB), +nonaclor and , '+DDE while the 116 second fraction, eluted with 20% DCM in hexane, contained hexachlorocyclohexane 117 isomers (HCHs), chlordane compounds (CHLs: oxychlordane, 118 chlordanes, 119 + and + and + +nonachlors) and DDTs ( , '+DDT, +DDE and +DDD) Quantification with external standards was carried out using a GC+ECD (Agilent 120 6890 series) equipped with an auto+injector (Agilent 7683 series) and a DB+1 fused 121 silica capillary column (0.25 mm i.d ì 0.25 àm film thickness ì 30 m length, J&W 122 Scientific, USA) The external standard for PCBs was an equivalent mixture of 62 PCB 123 congeners (BP+MS, Wellington Laboratories, Canada) A procedural blank was analysed 124 simultaneously with every series of five samples to check for interference and 125 contamination Recoveries throughout the procedure ( = 5) were 99.1 ± 5.5% for 126 DDTs, 95.7 ± 5.5% for PCBs, 94.1 ± 3.7% for HCHs, 92.6 ± 7.2% for HCB and 127 98.6 ± 6.5% for CHLs Relative standard deviations ( = 3) for the analysis of a pooled 128 milk sample were less than 15% for HCHs and less than 10% for other OCs Further 129 details on quantification and quality assurance have been described previously (Minh et 130 al., 2004) Concentrations were not recovery+corrected and were expressed on a lipid 131 weight basis unless otherwise specified 132 133 ! 134 The Wilcoxon rank sum test was used for assessing whether the contaminant levels 135 between groups were significantly different For this analysis, non+detectable levels 136 were set to zero Possible associations between levels of contaminants and 137 sociodemographic parameters were examined using multiple linear regressions 138 Compounds detected in less than 80% of the samples were not examined Non+ 139 detectable levels were set to half of the detection limit and then all levels were log+ 140 transformed (base 10) to bring the data distribution closer to normality The parameters 141 used as independent variables included age, body mass index (BMI), total lactation 142 time, and consumption of food from animal origin (total frequency for meat, fish and 143 dairy products) Parity was not included in the models due to a strong correlation with 144 lactation time (Spearman's ρ = 0.86, < 0.001) Parameters with a +value of more than 145 0.1 were removed from the model; those with < 0.05 were considered as having 146 significant relationship with level of the contaminant All calculations were performed 147 using the statistical software package R (R Foundation for Statistical Computing, 148 Vienna, Austria) version 2.9.2 149 First+order kinetics were assumed for both the depuration of OCs in individual 150 mother's milk through breastfeeding and the long+term decline of OCs in the population 151 This kinetic can be expressed using the following logarithmic equation: log10 152 = log10 – (Eq 1) 153 where 154 respectively and is the first+order decrease rate constant A decrease half+time can be 155 derived as and dec1/2 are the OC levels at the beginning and the end of the time interval , = log102 / 156 157 158 159 " ! # DDT compounds were the predominant organochlorine contaminants in Vietnamese 160 human breast milk The total levels of , '+DDT and its metabolites were an order of 161 magnitude higher than total PCB levels (Table 2) Other OCPs followed a pattern of 162 HCHs > HCB > CHLs The levels of CHLs were low; oxychlordane, 163 nonachlor were not detected in 27%, 39% and 21% of the samples, respectively These 164 results may reflect the usage pattern of OCPs in Vietnam, as the usage of HCB and 165 CHLs was very limited compared with other countries (Minh et al., 2004) + and + 166 As seen in Table 2, donors living in the city had significantly higher levels of OCs 167 than those living in suburban and rural areas Specifically, only , '+DDE, β+HCH and 168 several highly chlorinated PCB congeners (CB+138, +153 and +180) accumulated at 169 substantially higher levels in urban donors These compounds are the predominant and 170 more persistent of their respective groups Thus their elevated relative abundance 171 suggests a long accumulation history rather than a recent exposure Indeed, the 172 proportions of the main ingredients of pesticides technical mixtures, , '+DDT and α+ 173 HCH, within their groups of compounds were only 1.3%–8.5% (median 3.1%) and 174