The Vernix Caseosa is the Main Site of Dioxin Excretion in the Human Foetus

Dioxins are highly toxic to foetuses and prenatal exposure leads to adverse health effects; however, the metabolic pathways involved in dioxin excretion are poorly understood. We determined the dynamics of maternal-to-foetal dioxin transfer during normal pregnancy and how foetuses eliminate polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, and non-ortho polychlorinated biphenyls. Dioxin levels in maternal blood, cord blood, placenta, vernix caseosa, meconium, and amniotic fluid were analysed by high-resolution gas chromatography/mass spectrometry. The average levels of total dioxins, expressed as picograms of toxic equivalency quantity per gram of lipid and in parentheses, dioxin fraction, with maternal blood levels arbitrarily set as 100%, were as follows: maternal blood, 15.8 (100%); placenta, 12.9 (81.5%); cord blood, 5.9 (37.2%); vernix caseosa, 8.4 (53.2%); meconium, 2.9 (18.2%); and amniotic fluid, 1.5 (9.2%). Similar proportions were observed for each dioxin congener. Thus, the highest content of foetal dioxins was observed in the vernix caseosa, indicating that this is the major site of dioxin excretion in human foetuses.

The detection rate of dioxin congeners was 51% in maternal blood, 50% in umbilical cord blood, 62% in the placenta, 42% in the meconium, 53% in the vernix caseosa, and 13% in the amniotic fluid. PCDDs of either congener were not detected in nine amniotic-fluid samples. Similarly, PCDFs of either congener were not detected in two meconium and twelve amniotic-fluid samples. Regarding non-ortho PCBs, none of the congeners was detected in three amniotic-fluid samples. However, all other congeners were detectable.

Discussion
In the present study, we observed that the level of dioxins in cord blood was approximately 40% of that in maternal blood. However, the dioxin level in the foetal vernix caseosa was approximately 50% of the maternal blood level, indicating that the vernix caseosa constitutes an integral part of the dioxin excretion system in the human foetus. The weight and fat content of the vernix caseosa and meconium are approximately 5 g and 20-40% and 60-200 g and 2-3%, respectively. Therefore, the amount of fat of the vernix caseosa is approximately half that of the meconium. However, the level of dioxins per gram lipid in the vernix caseosa was more than twice that in meconium, indicating that dioxins are preferentially excreted into the vernix caseosa.
The vernix caseosa and meconium in the new-born are excellent depositories for persistent pesticide residues and hence can provide a historical record for prenatal exposure to these residues 9 . Therefore, these excrements are considered ideal matrices for analysing foetal exposure to various environmental toxins.
The vernix caseosa is a protective layer covering the foetus during the last trimester and is rich in cholesterol, free fatty acids, and ceramides 12 . Considering the highly lipophilic nature of dioxins [13][14][15] , it is logical to hypothesize that the inside-to-outside excretion of dioxins via lipid-rich epidermis, sebum, and the vernix caseosa may constitute a critical pathway enabling the foetus to lower the internal concentration of dioxins. Similar trans-epidermal excretion of dioxins has been previously described: high levels of dioxins were detected in the sebum of patients with Yusho disease and Yucheng disease (dioxin intoxication) 15 , and a study involving the Ukrainian President, who was exposed to highly concentrated tetrachlorodibenzo-p-dioxin, demonstrated the excretion of highly-concentrated dioxins in his sebum 14 . To the best of our knowledge, this study is the first to demonstrate that the vernix caseosa is one of the main sites of dioxin excretion in the human foetus.
We conclude that analysis of the vernix caseosa can detect antenatal foetal exposure to environmental toxins, specifically, heavy metals and pesticides. Vernix caseosa analysis is gaining acceptance in the scientific and medical communities as it comes with several advantages, such as provision of in utero record of exposure dioxins, non-invasiveness, and easy sample collection. However, as this pilot study was limited by its small sample number, no definite conclusions can be drawn, and future examination is necessary.
All experimental protocols were designed according to the Declaration of Helsinki's ethical principles, were performed in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects, and were approved by the Institutional Review Board of Kyushu University. Prior informed consent was obtained from all participants.
Analytical methods. After the extraction of lipids from the biological samples, the extracts were concentrated to dryness, and lipid contents were determined gravimetrically. PCDD, PCDF, and non-ortho PCB levels in the samples were analysed by high-resolution gas chromatography/mass spectrometry. The gas chromatograph used was an Agilent 7890 A (Agilent Technologies Inc., Palo Alto, CA, USA) equipped with an Autospec-Premier mass spectrometer (Waters Corp., Milford, MA, USA). Specifically, congeners of seven PCDDs, ten PCDFs, and four non-ortho PCBs were analysed according to previously published methods [16][17][18] . The dioxin detection limits per lipid weight were as follows: PCDDs and PCDFs, 0.3-2 pg/g lipid for umbilical cord blood, placenta, vernix caseosa, meconium, and amniotic fluid and 1-4 pg/g lipid for maternal blood; non-ortho PCBs, 0.3-0.6 pg/g lipid for placenta, 0.3-1 pg/g lipid for umbilical cord blood, placenta, vernix caseosa, meconium, and amniotic fluid and 10 pg/g lipid for maternal blood.
As the amounts of vernix caseosa and meconium obtained during delivery were limited, reproducibility tests using these specimens were not carried out in the present study. We identified that the recovery of the clean-up spike for each sample ranged from 50% to 120% according to the provisional manual for blood dioxin analysis issued by the Ministry of Health, Labour, and Welfare, Japan (2001). A reference serum sample was analysed for every batch, and analytical values fell within the predefined range.
To evaluate the accuracy and reliability of dioxin analysis, quality control studies on blood as a typical biological human tissue were completed once every year, and the results were compared against those obtained at three other laboratories in Japan. The average variation among the dioxin concentrations in human blood samples was considered acceptable if it was within 10%.
Because the toxic equivalent quantity (TEQ) has been proven a valuable measure in multiple studies on dioxins 19 , TEQs were calculated by multiplying the levels of individual congeners by their TEF value, as recommended by the World Health Organization in 2005 206 , and concentrations below the detection limit were designated as zero [18][19][20][21] .