This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Environment International 98 (2017) 46–53
Contents lists available at ScienceDirect
Environment International
j ourna l homepage: www.e lsev ie r .com/ locate /env int
Full length article
Glucuronide and sulfate conjugates of tetrabromobisphenol A (TBBPA):Chemical synthesis and correlation between their urinary levels andplasma TBBPA content in voluntary human donors
Ka-Lok Ho a, Ka-Ki Yuen a, Man-Shan Yau a, Margaret B. Murphy a, Yi Wan b,1, Bonnie M.-W. Fong c,d,Sidney Tam c, John P. Giesy a,b,e, Kelvin S.-Y. Leung d, Michael H.-W. Lam a,⁎a State Key Laboratory for Marine Pollution, Department of Biology and Chemistry, City University of Hong Kong, Hong Kong, Chinab Department of Biomedical Veterinary Sciences, Toxicology Centre, University of Saskatchewan, Canadac Division of Clinical Biochemistry, Queen Mary Hospital, Hong Kong, Chinad Department of Chemistry, Hong Kong Baptist University, Hong Kong, Chinae Department of Zoology and Center for Integrative Toxicology, Michigan State University, USA
⁎ Corresponding author at: State Key Laboratory for MBiology & Chemistry, City University of Hong Kong, TatKong, China.
E-mail address: [email protected] (M.H.-W. La1 Present address: Laboratory for Earth Surface Pro
Environmental Sciences, Peking University, Beijing, China
Article history:Received 27 June 2016Received in revised form 16 September 2016Accepted 22 September 2016Available online 4 October 2016
3,3′,5,5′-Tetrabromobisphenol-A (TBBPA) is an important brominated flame retardant in epoxy, vinyl esters andpolycarbonate resins. Previous studies have already shown the occurrence of its Phase II metabolites, TBBPA-glu-curonide and sulfate conjugates, in human urine, after oral administration of TBBPA. The main objective of thiswork is to examine correlations among level of TBBPA in human blood and those of its Phase II metabolites inhuman urine. Four water-soluble TBBPA conjugates were synthesized, purified and characterized. An analyticalprotocol using solid-phase extraction and liquid chromatography–electrospray tandem mass spectrometry(SPE-LC-MS/MS) quantification was developed for the simultaneous analysis of these glucuronide and sulfateconjugates in human urine samples. TBBPA and its Phase II metabolites in paired human plasma and urine sam-ples collected randomly from 140 voluntary donors in Hong Kong SAR, China, were determined. One or moreTBBPA conjugates were detected in all of the urine samples, with concentration ranging from 0.19 to 127.24 μgg−1-creatinine. TBBPAwas also quantified in N85% of the plasma and urine samples. Strong correlationswere ob-served between TBBPA content in plasma and the total amount of TBBPA-related compounds in urine.
Keywords:Tetrabromobisphenol-AGlucuronide and sulfate conjugatesBody fluidsHuman plasmaHuman urine
1. Introduction
Environmental endocrine disrupting chemicals (EDCs) are thosechemical species that occur in the various environmental compartmentsthat can affect normal hormonal signaling in living organisms(Nicolopoulou-Stamati et al., 2001). Their potential risks to environ-mental and human health have aroused much public concerns (Norrisand Carr, 2006). 3,3′,5,5′-Tetrabromobisphenol-A (TBBPA) is a syntheticchemicalwidely used as a brominated flame retardant, and is, therefore,commonly found in a wide variety of daily appliances (Birnbaum andStaskal, 2004, Shi et al., 2009a). Recent studies have revealed its endo-crine disrupting impacts on the reproduction and the functioning ofthyroid hormones of laboratory animals (Saegusa et al., 2009, van der
arine Pollution, Department ofChee Avenue, Kowloon, Hong
m).cesses, College of Urban and.
Ven et al., 2008). Toxicological and human exposure data that wehave gathered thus far warrant our concern that TBBPA may have al-ready exerting their adverse impacts on human health (Kim and Oh,2014). As Southern China is one of the major manufacturing hubs forconsumer products in the world, there is much concern about environ-mental contamination and public health risk posed by EDCs in the re-gion (Feng et al., 2012, He et al., 2010, Shi et al., 2009a, Tan et al.,2016, Tang et al., 2015, Wang et al., 2015).
At the moment, the most commonly adopted approach for the esti-mation of population exposure to selected environmental contaminantsis the direct quantification of their levels in human tissues, mainlywhole blood/serum (Cariou et al., 2008, Dirtu et al., 2008, Dirtu et al.,2010, Fujii et al., 2014a, Fujii et al., 2014b, Jakobsson et al., 2002, Kimand Oh, 2014, Nagayama et al., 2000, Shi et al., 2013b, Thomsen et al.,2001, Thomsen et al., 2002), and breast milk (Abdallah and Harrad,2011; Carignan et al., 2012; Cariou et al., 2008; Fujii et al., 2014b;Kang et al., 2015; Lankova et al., 2013; Nakao et al., 2015; Shi et al.,2009b; Shi et al., 2013a, 2013b) from voluntary donors recruited fromthe population. Other less commonly used tissues for exposure estima-tion include adipose tissue (Cariou et al., 2008, Johnson-Restrepo et al.,
47K.-L. Ho et al. / Environment International 98 (2017) 46–53
2008), and hair (Pellizzari et al., 1978). Nevertheless, sampling humantissues for contaminant analysis is intrusive and difficult to achieve ona large scale, especially for healthy subjects who are not hospitalizedor required to go through clinical diagnostic tests. Also, this approachcan only provide a snapshot view of the exposure situation, unless along-term and extensive sampling arrangement is available, which de-mands huge amount of resources to sustain. Sampling of breast mikecan be considered a non-invasive operation, but is very much that re-stricted to lactating women within a relatively narrow age distribution(Landrigan et al., 2002).
Because of their high molecular mass, excretion via urine only con-stitutes a minor elimination route for TBBPA and its metabolites inhuman (Szymańska et al., 2001, Kuester et al., 2007). Nevertheless, glu-curonide and sulfate conjugated Phase II metabolites of TBBPA havebeen quantified in human urine (Fini et al., 2012, Schauer et al., 2006).In vitro studies on the biotransformation of TBBPA carried out on rat he-patocytes have also revealed the production of TBBPA sulfate and glucu-ronide conjugates (Nakagawa et al., 2007). Thus, it is deemedworthy ofexamining correlations of these glucuronide and sulfate conjugates ofTBBPA in human urine with TBBPA exposure. In this context, we reportthe synthesis and purification of all the four common Phase II metabo-lites of TBBPA (TBBPAmono-/di-glucuronide andmono-/di-sulfate con-jugates), and their use as authentic standards for the development of ananalytical protocol for their determination in human urine. We havealso studied the stability of these TBBPA Phase II metabolites inhuman urine and derived means to preserve them for quantitative de-termination. Finally, correlations among their levels in human urineand that of TBBPA in human blood are investigated.
2. Materials and methods
2.1. Safety precautions
Extra precaution was practiced in the handling of human blood andurine samples. Double latex gloves, facemasks and eye-protection gog-gles were worn all the time during their handling, spiking and transfer-al. All the spent samples after analysis were collected in a separatedclose-lipped container with proper clinical waste labels. These spentsamples and all the used personal protection itemswere treated as clin-ical wastes and were collected and disposed of in accordance with the“Code of Practice for the Management of Clinical Waste” issued by theEnvironmental Protection Department of the Hong Kong SARGovernment.
2.2. Sample collection
Human studies were performed in accordance with the guidelinesand approval of the research ethics committee of City University ofHong Kong. Parallel human plasma and urine samples from 140 volun-tary donorswere collected fromApril to August 2012, by registered doc-tors and nurse, at Queens Mary Hospital, Hong Kong SAR. Of the 140donors, 66weremale and 74were female. The age range of these volun-teers was from 18 to 96 year-old, with the mean age = 47.1 ± 18.2year-old. These volunteers were subdivided into different age groupsfor comparison as following: age 18 to 25 (n = 17); age 26 to 35(n = 18); age 36 to 45 (n = 44); age 46 to 55 (n = 24); age 56 to 65(n = 15); age 66 to 80 (n = 13) and age N 80 (n = 9).
Whole blood sampleswere collected using the standard phlebotomytechnique in vacutainer tubes containing sodium heparin anticoagulant(Vacuette, Greiner bio-one, GmbH, Austria). The blood was then centri-fuged at 1500 ×g for 25 min. Plasma was removed from the top of thetube. Urine samples were collected in 100 mL sterilized glass bottlesand stored at −80 °C, within 15 min after sampling, until being ana-lyzed. Urine sample from each donor was subdivided into three repli-cate samples before low temperature storage. All samples werecarefully labeled and documented. Upon analysis, samples were
thawed, and 10 mL of each sample was taken for creatinine content de-termination. Creatinine determination was carried out by a kinetic col-orimetric assay based on the modified Jaffe method (Bonsnes andTaussky, 1945) using the Roche Modular System (Roche Diagnostics,IN, USA), with an analytical range between 360 and 57,500 mmol L−1.
2.3. Synthesis, purification and characterization of TBBPA glucuronides andsulfate conjugates
The general synthetic routes for the TBBPA glucuronide and sulfateconjugates are outlined in Scheme S1 of the Supporting information.Detail synthetic and purification procedures and characterization dataare also given in the Supporting information.
2.4. Sample extraction and cleanup
2.4.1. TBBPA in human plasmaLiterature method (Dirtu et al., 2008, Hovander et al., 2000), with
slight modifications, was adopted for the extraction of TBBPA inhuman plasma. 13C12-TBBPA (1 ng) was spiked to 1 mL of human plas-ma sample. The spiked plasma sample was allowed to stand at roomtemperature for 10 min before 50 μL of concentrated hydrochloric acid(37%), Milli-Q water (2 mL) and isopropanol (3 mL) were added. Themixture was then extracted by hexane/MTBE (3 × 5 mL, 1:1, v/v). Theorganic fractions were combined and partitionedwith a 1% KCl solution(3 mL) followed by evaporation to dryness under a gentle steam of ni-trogen. Lipid content of the plasma sample was determined gravimetri-cally. The residue was re-dissolved in hexane (4mL) and partition withpotassiumhydroxide (2mL, 0.5M in 50% ethanol) to ionize thephenolicanalytes. Neutral compoundswere separated by hexane (2× 4mL). Theaqueous layer was acidified by hydrochloric acid (2mL, 0.5M), then thephenolic compoundswere extracted by hexane/MTBE (2 × 4mL, 9:1, v/v). Phenolic fractionwas evaporated to dryness under a gentle streamofnitrogen and reconstituted in 1mLof 5% acetone in hexane. Themixturewas subjected to a SPE clean-up using a Sep.-Pak Florisil cartridge previ-ously conditioned by DCM/MeOH (6 mL, 4:1, v/v) and 5% acetone inhexane (6 mL) at the flow rate of 1 drop/s. The cartridge then washedby 5% acetone in hexane (6 mL). The cartridge was the dried under re-duced pressure and the BPA and BPA-d16 were eluted by DCM/MeOH(10 mL, 4:1, v/v). The eluate was evaporated to dryness under a gentlestream of nitrogen. Analytes were reconstituted in 50 μL isooctane con-taining 5 ng pyrene-d10 as an internal standard and the phenolicanalytes were derivatized by 50 μL BSTFA with 1% TMCS at 70 °C foran hour.
2.4.2. TBBPA in human urineLiterature method (Ho et al., 2016), with slight modifications, was
adopted from the extraction of TBBPA from human urine. Briefly,13C12-TBBPA (1 ng)was spiked in 5mL of human urine. The spiked sam-ple was then allowed to stand at room temperature for 10 min. Then,50 μL of formic acid was added, followed by extraction with ethyl ace-tate (3 × 5 mL). The organic fractions were combined and evaporatedto dryness under a gentle steam of nitrogen. This was followed by re-constitution in 5% acetone in hexane (1 mL), and SPE clean-up with aSep.-Pak Florisil cartridge previously conditioned by DCM/MeOH(6 mL, 4:1, v/v) and 5% acetone in hexane (6 mL) at a rate of 1 drop/s.The cartridge was then eluted by 5% acetone in hexane (6mL). The car-tridge was the dried under reduced pressure and the TBBPA and 13C12-TBBPA were eluted by DCM/MeOH (10 mL, 4:1, v/v). The eluate wasevaporated to dryness under a gentle stream of nitrogen. Analytes wasreconstituted in 50 μL isooctane containing 5 ng pyrene-d10 as an inter-nal standard and the phenolic analytes were derivatized by 50 μL BSTFAwith 1% TMCS at 70 °C for an hour.
48 K.-L. Ho et al. / Environment International 98 (2017) 46–53
2.4.3. TBBPA glucuronide and sulfate conjugates in human urineLiterature method (Ho et al., 2016), with slight modifications, was
adopted from the extraction of the highly water-soluble TBBPA conju-gates from human urine. 13C12-TBBPA (10 ng) was spiked in 5 mL ofhuman urine. The spiked sample was then allowed to stand at roomtemperature for 10min. Then,wasmixedwith 2M sodium acetate buff-er (5 mL, pH = 5) and then loaded at the flow rate of 1 drop/s onto anOasis® WAX cartridge previously conditioned with 5 mL of methanol,then 5 mL of Milli-Q water, then 5 mL of 2 M sodium acetate buffer.The cartridge was then washed by a 2 M sodium acetate buffer (5 mL,pH = 5), Milli-Q water (5 mL), and 2% acetic acid in methanol (5 mL).The sorbent was then dried under reduced pressure and the
H3
H3
H4
H4
Fig. 1. 1H-1H ROESY 2D-NMR spectra of TBBPA mono-glucuronide (upper) and TBBPA di-gglucuronide rings indicates the β-conformation of the rings.
glucuronide and sulfate conjugates were eluted by 5% NH4OH in meth-anol (10 mL). The elute was evaporated to dryness under a gentlestream of nitrogen at 50 °C and was then reconstituted in a mixture ofacetonitrile and water (1:1, 100 μL) containing 13C12-tetrachlorobisphenol A (13C12-TCBPA, 50 ng). The samplewasfinally an-alyzed by HPLC-MS/MS.
2.5. Quality control (QC), quality assurance (QA) and statistical analysis
Details on QA/QC and statistical analysis for this study are given inthe Supporting Information.
H4
H5, H
6, H7
H5
H6,H
7, H8
lucuronide (lower). The absence of cross signal between the C1 and C2 carbons of the
Table 1LC-MS/MS transitions and MS parameters of TBBPA Phase II conjugates (upper) and GC–MS/MS transitions and MS parameters of TBBPA analysis (lower) adopted in this study.
49K.-L. Ho et al. / Environment International 98 (2017) 46–53
3. Results and discussion
3.1. Synthesis and purification of the glucuronide and sulfate conjugates ofTBBPA
Mono- and di-sulfate conjugate of TBBPA were prepared accordingto the previous published method with few modifications (Burkhardtand Lapworth, 1926). The excess chlorosulfonic acid cannot be neutral-ized by inorganic strong base like KOH or K2CO3 because the sulfateester linkage will be cleaved in the presence of inorganic bases. Instead,the products have to be precipitated by triethylamine and subjected topurification by HPLC. The sulfate conjugates were in their triethylaminesalts. D-(+)-glucuronic acid γ-lactone was protected by acetic anhy-dride in the presence of perchloric acid and then the activated C1 posi-tion was brominated by HBr (33%) in glacial acetic acid. The sugarmoiety was then linked to TBBPA in an appropriate mole ratio to obtainTBBPA mono-2,3,4-tri-O-acetyl-β-D-glucopyranosiduronic acid methylester and TBBPA di-2,3,4-tri-O-acetyl-β-D-glucopyranosiduronic acidmethyl ester. The protected sugar moieties were then deprotected byexcess KOH in methanol to give the desire products. Such a syntheticroute is different from that adopted in our previous study for the syn-thesis of bromophenol glucuronides (Ho et al., 2012). In fact, we havetried that synthetic route, but observe a cleavage of the ether linkagesbetween TBBPA and the sugar moieties under TEMPO mediated oxida-tion of primary alcohols. Stereochemistry of our synthesized TBBPAmono-/di glucuronides was confirmed by 1H-1H ROESY 2D-NMR(Fig. 1) to sustain the β-confirmation, which is the same as natural glu-curonide metabolites in human (Mulder, 1992).
Table 2Performance of the SPE-LC-MS/MS analytical protocol for the determination of the TBBPAglucuronide and sulfate conjugates, and the SPE-GC–MS/MS analytical protocol for the de-termination of the TBBPA in artificial urine matrix.
Previous studies on the pharmacokinetics and toxicokinetics ofTBBPA in human andmammalian animal models have already reportedthe occurrence of urinary TBBPA glucuronide and sulfate conjugates(Fini et al., 2012, Hakk et al., 2000, Nakagawa et al., 2007, Knudsen etal., 2014, Riu et al., 2011, Schauer et al., 2006, Zalko et al., 2006). Yet,those Phase II metabolites of TBBPA have only been qualitatively identi-fied. To the best of our knowledge, no information is available on theirrelative abundance in human urine, not to mention any correlationwith their parent TBBPA compound in human blood. Thus, we set outto develop an analytical protocol for the qualitative and quantitative de-termination of TBBPA and its Phase II metabolites in human urine.Table 1 tabulates all the Multiple Reaction Monitoring (MRM) transi-tions adopted for the identification and quantification of TBBPA, andTBBPA glucuronide and sulfate conjugates in this study. Good linearLC-MS/MS responses were obtained for different concentrations ofmono-/di-glucuronide and sulfate substituted conjugates of TBBPA cov-ered the concentration range of 0.5 to 1000ngmL−1. The correlation co-efficients of the best-fitted line, r2, were in the range of 0.9987 to 0.9997for the four different substituted glucuronide and sulfate conjugates ofTBBPA. In addition, linearity of GC–MS/MS responses to parent TBBPAand 13C12-TBBPA was analyzed for the range from 0.5 to 1000 ngmL−1. The correlation coefficients, r2, of the best-fitted linesare 0.9993 and 0.9994 for TBBPA and 13C12-TBBPA respectively. Repeat-ability and analyte recovery were evaluated by the consecutive analysisof seven independent artificial urine samples spiked with 0.5 ng mL−1
of each of the conjugates. Table 2 summaries the optimize perfor-mance of the analytical protocol for the determination of theTBBPA Phase II conjugates. The method detection limits (MDLs) forall the conjugates were ≤6.4 ng g−1 creatinine. Finally, robustnessof the analytical protocol was assessed by analyzing the same spikedsample once a week for seven weeks. No variation in the relative re-tention time (RRT) of the conjugates whatsoever was observed. Fluc-tuations in the corresponding chromatographic peak areas of theconjugates were within 5%.
3.3. Stability of TBBPA glucuronide and sulfate conjugate in human urine
Stability of the synthesized TBBPA Phase II metabolites under vari-ous preservation conditions was assessed, and results are shown inFig. 2. Similar to those Phase II metabolites of bromophenols andbisphenol-A, the glucuronide and sulfate conjugates of TBBPA are ratherstable at low temperature (Ho et al., 2012, Ho et al., 2016). They remainunchanged in 2 weeks and have only decreased about one fifth in30 days. On the other hand, neither addition of formaldehyde, nor sodi-um azide, was able to suppress their rather rapid degradation. Thus, lowtemperature storage (−80 °C) was adopted to preserve human urinesamples in this study.
3.4. Levels of TBBPA in human plasma and human urine samples
TBBPA was quantified in ca. 93% of the plasma samples of the 140voluntary donors. The geometric mean content was 3.65 ng g−1 l.w,
0 5 10 15 20 25 30
0
20
40
60
80
100
120(a)
Rel
ativ
eR
esp
on
se(%
)
Day
0 5 10 15 20 25 30
0
20
40
60
80
100
120
(b)
Rel
ativ
eR
esp
on
se(%
)
Day
0 5 10 15 20 25 30
0
20
40
60
80
100
120(c)
Rel
ativ
eR
esp
on
se(%
)
Day
0 5 10 15 20 25 30
0
20
40
60
80
100
120
Rel
ativ
eR
esp
on
se(%
)
Day
Non-Preserved Human Urine
-20oCPreserved
NaN3
Preserved
Formaldehyde Preserved
(d)
Non-Preserved Human Urine
-20oCPreserved
NaN3
Preserved
Formaldehyde Preserved
Non-Preserved Human Urine
-20oCPreserved
NaN3
Preserved
Formaldehyde Preserved
Non-Preserved Human Urine
-20oCPreserved
NaN3
Preserved
Formaldehyde Preserved
Fig. 2. Stability of TBBPA conjugates under chemical and physical means of preservation: (a) TBBPAmono-glucuronide, (b) TBBPAmono-sulfate, (c) TBBPA di-glucuronide, and (d) TBBPAdi-sulfate.
50 K.-L. Ho et al. / Environment International 98 (2017) 46–53
(95% confidence interval: 3.44–3.88 ng g−1 l.w). This plasma concentra-tion of TBBPA is quite comparable to that revealed in other studies onthe general population of China and Japan (Nagayama et al., 2000, Shiet al., 2013b), and workers (Jakobsson et al., 2002, Thomsen et al.,2001). On the other hand, it is higher than those in the general popula-tion of European countries (Jakobsson et al., 2002, Thomsen et al.,2002), except the maternal serum and cord serum of French mothers(Cariou et al., 2008), and plasma samples collected in Belgium (Dirtuet al., 2008). High concentration of TBBPA has also been detected in Ko-rean infant-mother paired serum samples (Kim and Oh, 2014). Such alevel of TBBPA in human plasma collected in Hong Kong SAR is probablydue to the high production and consumption rate of TBBPA in the PearlRiver Delta region of Southern China. Previous studies in Southern China
Table 3Concentrations of TBBPA and TBBPA mono-/di-glucuronide and sulfate conjugates in paired hSouth China (n = 140).
Total (n = 140) Male (n = 6
GMa (95% CI)b Min–max % of detection GMa (95% CI
Human plasma samplesTBBPA 3.65
(3.44–3.88)N.D. – 7.49 92.9 3.55
(3.25–3.88)
Human urine samplesTBBPA 0.19
(0.13–0.29)N.D. – 88.41 88.6 0.14
(0.08–0.25)TBBPAmono-glucuronide
1.69(1.20–2.38)
N.D. – 118.06 86.4 1.24(0.80–1.21)
TBBPA di-glucuronide 1.74(1.33–2.26)
N.D. – 95.57 99.3 1.32(0.87–2.01)
TBBPA mono-sulfate 0.63(0.51–0.78)
N.D. – 26.86 82.1 0.48(0.36–0.64)
TBBPA di-sulfate 0.89(0.69–1.14)
N.D. – 25.80 82.9 0.81(0.55–1.21)
N.D.: not detected.a GM: geometric mean.b CI: confidence interval.
have already detected high TBBPA content in bird samples collectednear e-waste recycling sites (He et al., 2010). There is no obvious differ-ence in the plasma TBBPA content between male and female donors(natural-logarithmic transformed, P = 0.248). Also, there is no signifi-cant difference in the median level of TBBPA in human plasma amongage groups (natural-logarithmic transformed, P = 0.947). TBBPA wasalso quantified in N85% of urine samples of the voluntary donors. Itsgeometric mean level was 0.19 μg g−1 creatinine (95% confidence inter-val: 0.13–0.29 μg g−1 creatinine) (Table 3). No statistically significantdifference is observed between male and female donors (natural-loga-rithmic transformed, P = 0.368). There is also no statistical differencein the urinary TBBPA content among age groups (natural-logarithmictransformed, P = 0.109).
uman plasma (ng mL−1) and human urine samples (μg g−1 creatinine) from Hong Kong,
6) Female (n = 74)
)b Min–max % of detection GMa (95% CI)b Min–max % of detection
Table 4Correlation coefficients between parent TBBPA in human plasma and its Phase II metabo-lites in urine.
Total (n =140)
Male (n =66)
Female (n =74)
r P r P r P
Urinary content (in terms of mass concentration)ln[parent TBBPA] 0.547 b0.05 0.534 b0.05 0.586 b0.05lnΣ[TBBPA Phase II metabolites] 0.626 b0.05 0.668 b0.05 0.599 b0.05ln{[parent TBBPA] + Σ[TBBPAPhase II metabolites]}
0.886 b0.005 0.873 b0.05 0.897 b0.05
Urinary content (in terms of molar concentration)ln[TBBPA-related compounds] 0.913 b0.005 0.901 b0.05 0.927 b0.05
51K.-L. Ho et al. / Environment International 98 (2017) 46–53
3.5. Levels of TBBPA glucuronide and sulfate conjugates in human urinesamples
Concentrations of TBBPA glucuronide and sulfate conjugates inhuman urine samples are summarized in Table 3. At least one of fourTBBPAPhase IImetaboliteswas detected in the urine samples of the vol-untary donors. TBBPA di-glucuronidewas themost frequently detected.Levels of urinary TBBPAmono-glucuronide anddi-glucuronide are com-parable, with geometric means of 1.69 μg g−1 creatinine (95% confi-dence interval: 1.20–2.38) and 1.74 μg g−1 creatinine (95% confidenceinterval: 1.33–2.66 μg g−1 creatinine), respectively, which are muchhigher than those of TBBPA mono- & di-sulfate. Of the four TBBPA con-jugates, only the urinary TBBPA di-glucuronide showed statistically dif-ference in concentration betweenmale and female (natural-logarithmictransformed, P=0.012). Higher concentration of TBBPA di-glucuronidewas found in women (median= 0.883 μg g−1 creatinine) compared tomen (median = 0.368 μg g−1 creatinine). However, in term of age,there is no statistically difference observed for all the TBBPA Phase IIconjugates.
3.6. Correlation studies
We employed Pearsons Product Moment to examine relationshipsamong urinary levels (natural-logarithmic transformed) of TBBPA andits conjugates and that of TBBPA in the corresponding blood plasma ofvoluntary donors. Results are tabulated in Table 4 and shown in Figs. 3and 4. In terms of mass concentration of TBBPA, only moderate correla-tion, r = 0.547, is found between blood plasma and urine. Correlationbetween the mass concentration of plasma TBBPA and that of all thequantified TBBPA Phase II conjugates in urine is also not very substantial(r = 0.626). Indeed, a much improved correlation, r = 0.886, is foundbetween the mass concentration of plasma TBBPA and that of urinary“TBBPA-related compounds” (i.e. the parent TBBPA and all the TBBPAglucuronide and sulfate conjugates).
Fig. 3. Correlation of plasma level of TBBPAwith: (i) TBBPA content in human urine (left); (ii) tocontent of “TBBPA-related compounds” (i.e. both the parent TBBPA and its Phase II conjugatelogarithmic transformed.
Expressing the content of urinary metabolites of organic contami-nants in terms of their mass concentration may not be able to truly re-flect the quantity that had been transformed from their parentcompounds as the resulting conjugation groups affect their molecularmasses. a survey of environmental exposure of pregnant women tobisphenol A (BPA), Gerona et al. (2016) converted the mass concentra-tions of BPA mono-glucuronide and sulfate conjugates into their “BPA-equivalent” mass concentrations using the ratios of the molecularmass of BPA to that of its corresponding conjugates as theweighing fac-tors. Thus, to further explore the correlation between TBBPA in humanplasma and its metabolites in human urine, we calculated the molarconcentration (i.e. in mol mL−1 for plasma content and mol g−1-creat-inine for urinary content) of TBBPA and its Phase II conjugates in theplasma and urine samples (Eq. 1):
Molar concentration of contaminant
¼ mass concentration of contaminantmolecular mass of contaminant
ð1Þ
Molecular mass of TBBPA (543.9 g mol−1), TBBPA mono-glucuro-nide (703.0 g mol−1), TBBPA di-glucuronide (894.1 g mol−1), TBBPAmono-sulfate (triethylammonium salt) (725.1 g mol−1) and TBBPAdi-sulfate (bis-triethylammonium salt) (906.4 g mol−1) were used totransform themass concentration of TBBPA and its glucuronide and sul-fate conjugates into their corresponding molar concentration in theplasma andurine samples.Re-examining the relationship between plas-ma TBBPA content and that of all “TBBPA-related compounds” in urinereveals an even stronger correlation with r = 0.913 (Fig. 4).
Previous studies have revealed that TBBPA is efficiently metabolizedin the liver by glucuronyl- and sulfotransferases, which are, in turn,eliminated mainly into bile (Schauer et al., 2006, Kuester et al., 2007).Because of the resulting enterohepatic circulation, systemic bioavail-ability of TBBPA in mammalian models, including human, is not high.Half-life of TBBPA in human plasma is approximately 2 days (Hagmaret al., 2000, Schauer et al., 2006). Thismay explain the relatively poor as-sociation between the plasma TBBPA content and the urinary levels ofits Phase II conjugates. Nevertheless, this study reveals that the com-bined content of TBBPA and all of its Phase II conjugates shows muchimproved correlation with the plasma TBBPA level. In view of the rela-tively complex metabolism and elimination pathways of TBBPA inhuman, it is deemed too early to judge whether such a urinary contentof “TBBPA-related compounds” can be a reliable population exposuremarker for TBBPA. Further works to study their relationships withTBBPA in other body fluids, e.g. bile, and tissues are definitely needed.
In conclusion, all the four common Phase II metabolites of TBBPAhave been successfully synthesized, purified and characterized as au-thentic standards for analytical purposes. An SPE-LC-MS/MS analyticalprotocol has been developed for their quantitative determination inhuman urine. Through the analysis of paired blood plasma and urinesamples from 140 voluntary donors in Hong Kong SAR, China, a good
tal content of all the four Phase II conjugates of TBBPA in human urine (middle); (iii) totals) in human urine (right). Levels of TBBPA and its Phase II conjugates have been natural-
Fig. 4. Correlation of molar concentrations (natural-logarithmic transformed) TBBPA inplasma with that of the “TBBPA-related compounds” in urine of voluntary donors.
52 K.-L. Ho et al. / Environment International 98 (2017) 46–53
correlation has been revealed between plasma level of TBBPA and theurinary content of all the “TBBPA-related compounds” (parent TBBPAand all of its four Phase II conjugates). We hope that these results areuseful for the establishment of more convenient population exposuremarkers for TBBPA.
Acknowledgement
This work is support by the State Key Laboratory for Marine Pollu-tion, City University of Hong Kong, and a GRF grant CityU 160513from the Research Grants Council of the Hong Kong Special Administra-tive Region, China. Prof. Giesy was supported by the program of 2012“Great Concentration Foreign Experts” (#GDW20123200120) fundedby the State Administration of Foreign Experts Affairs, the P.R. China toNanjing University and the Einstein Professor Program of the ChineseAcademy of Sciences. He was also supported by the Canada ResearchChair program, a Visiting Distinguished Professorship in the Depart-ment of Biology and Chemistry and State Key Laboratory in Marine Pol-lution, City University of Hong Kong.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.envint.2016.09.018.
References
Abdallah, M.A.E., Harrad, S., 2011. Tetrabromobisphenol-A, hexabromocyclododecane andits degradation products in UK human milk: relationship to external exposure. Envi-ron. Int. 37, 443–448.
Birnbaum, L.S., Staskal, D.F., 2004. Brominated flame retardants: cause for concern. Envi-ron. Health Perspect. 112, 9–17.
Bonsnes, R.W., Taussky, H.H., 1945. On the colorimetric determination of creatinine by theJaffe reaction. J. Biol. Chem. 158, 581–591.
Burkhardt, G.N., Lapworth, A., 1926. XCV—arylsulphuric acids. J. Chem. Soc. 684–690.Carignan, C.C., Abdallah, M.A.E., Wu, N., Heiger-Bernays, W., McClean, M.D., Harrad, S.,
Webster, T.F., 2012. Predictors of tetrabromobisphenol-A (TBBP-A) andhexabromocyclododecanes (HBCD) in milk from Boston mothers. Environ. Sci.Technol. 46, 12146–12153.
Cariou, R., Antignac, J.P., Zalko, D., Berrebi, A., Cravedi, J.P., Maume, D., Marchand, P.,Monteau, F., Riu, A., Andre, F., Le bizec, B., 2008. Exposure assessment of Frenchwomen and their newborns to tetrabromobisphenol-A: occurrence measurementsin maternal adipose tissue, serum, breast milk and cord serum. Chemosphere 73,1036–1041.
Dirtu, A., Roosens, L., Geens, T., Gheorghe, A., Neels, H., Covaci, A., 2008. Simultaneous de-termination of bisphenol a, triclosan, and tetrabromobisphenol a in human serumusing solid-phase extraction and gas chromatography-electron capture negative-ion-ization mass spectrometry. Anal. Bioanal. Chem. 391, 1175–1181.
Dirtu, A., Jaspers, V.B., Cernat, R., Neels, R., Covaci, A., 2010. Distribution of PCBs, their hy-droxylated metabolites, and other phenolic contaminants in human serum from twoEuropean countries. Environ. Sci. Technol. 44, 2876–2883.
Feng, A.H., Chen, S.J., Chen, M.Y., He, M.J., Luo, X.J., Mai, B.X., 2012. Hexabromocyclododecane(HBCD) and tetrabromobisphenol A (TBBPA) in riverine and estuarine sediments of thePearl River Delta in southern China, with emphasis on spatial variability in diastereoiso-mer- and enantiomer-specific distribution of HBCD. Mar. Pollut. Bull. 64, 919–925.
Fini, J.-S., Riu, A., Debrauwer, L., Hillenweck, A., Mével, S., Chebolleau, S., Boulahtouf, A.,Palmier, K., Balaguer, P., Cravedi, J.-P., Demeneix, B.A., Zalko, D., 2012. Parallel bio-transformation of Tetrabromobisphenol a in Xenopus laevis and mammals: Xenopusas a model for endocrine perturbation studies. Toxicol. Sci. 125, 359–367.
Fujii, Y., Harada, K.H., Hitomi, T., Kobayashi, H., Koizumi, A., Haraguchi, K., 2014a. Tempo-ral trend and age-dependent serum concentration of phenolic organohalogen con-taminants in Japanese men during 1989–2010. Environ. Pollut. 185, 228–233.
Fujii, Y., Nashimura, E., Kato, Y., Harada, K.H., Koizumi, A., Haragushi, K., 2014b. Dietary ex-posure to phenolic and methoxylated organohalogen contaminants in relation totheir concentrations in breast milk and serum in Japan. Environ. Int. 63, 19–25.
Gerona, R.R., Pan, J., Zota, A.R., Schwartz, J.M., Friesen, M., Taylor, J.A., Hunt, P.A., Woodruff,T.J., 2016. Directmeasurement of bisphenol A (BPA), BPA glucuronide and BPA sulfatein a diverse and low-income population of pregnant women reveals high exposure,with potential implications for previous exposure estimates: a cross-sectionalstudy. Environ. Health 15, 50.
Hagmar, L., Sjödin, A., Höglund, P., Thuresson, K., Rylander, L., Bergman, A., 2000. Biolog-ical half-lives of polybrominated diphenyl ethers and tetrabromobisphenol A in ex-posed workers. Organohalogen Compd. 47, 198–201.
Hakk, H., Larsen, G., Bergman, Å., Orn, U., 2000. Metabolism, excretion and distribution ofthe flame retardant tetrabromobisphenol-A in conventional and bile-duct cannulatedrats. Xenobiotica 30, 881–890.
He, M.J., Luo, X.J., Yu, L.H., Liu, J.A., Zhang, X.L., Chen, S.J., Chen, D., Mai, B.X., 2010.Tetrabromobisphenol-A and hexabromocyclododecane in birds from an E-waste re-gion in South China: influence of diet on diastereoisomer- and enantiomer-specificdistribution and trophodynamics. Environ. Sci. Technol. 44, 5748–5754.
Ho, K.L., Murphy, M.B., Yi, W., Fong, B.M.W., Tam, S., Giesy, J.P., Leung, K.S.Y., Lam, M.H.W.,2012. Synthesis and characterization of bromophenol glucuronide and sulfate conju-gates for their direct LC-MS/MS quantification in human urine as potential exposuremarkers for polybrominated diphenyl ethers. Anal. Chem. 84, 9881–9888.
Ho, K.L., Yuen, K.K., Yau, M.S., Murphy, M.B., Wan, Y., Fong, B.M.W., Tam, S., Giesy, J.P.,Leung, K.S.Y., Lam, M.H.W., 2016. Glucuronide and Sulfate Conjugates of BisphenolA: Chemical Synthesis and Correlation Between Their Urinary Levels and PlasmaBisphenol A Content in Voluntary Human Donors (Unpublished).
Hovander, L., Athanasiadou, M., Asplund, L., Jensen, S., Wehler, E.K., 2000. Extraction andcleanup methods for analysis of phenolic and neutral organohalogens in plasma.J. Anal. Toxicol. 24, 696–703.
Jakobsson, K., Thuresson, K., Rylander, L., Sjödin, A., Hagmar, L., Bergman, Å., 2002. Expo-sure to polybrominated diphenyl ethers and tetrabromobisphenol A among comput-er technicians. Chemosphere 46, 709–716.
Johnson-Restrepo, B., Adams, D.H., Kannan, K., 2008. Tetrabromobisphenol A (TBBPA) andhexabromocyclododecanes (HBCDs) in tissues of humans, dolphins, and sharks fromthe United States. Chemosphere 70, 1935–1944.
Kang, H., Wang, X., Zhang, Y., J., W., 2015. Simultaneous extraction of bisphenol A andtetrabromobisphenol A from milk by microwave-assisted ionic liquidmicroextraction. RSC Adv. 5, 14631–14636.
Kim, U.-J., Oh, J.-E., 2014. Tetrabromobisphenol A and hexabromocyclododecane flame re-tardants in infant-mother paired serum samples, and their relationships with thyroidhormones and environmental factors. Environ. Pollut. 184, 193–200.
Knudsen, G.A., Sanders, J.M., Sadik, A.M., Birnbaum, L.S., 2014. Disposition and kinetics oftetrabromobisphenol A in female Wistar Han rats. Toxicol. Rep. 1, 214–223.
Kuester, R.K., Sólyom, A.M., Rodriguez, V.P., Sipes, G., 2007. The effects of dose, route, andrepeated dosing on the disposition and kinetics of tetrabromobisphenol A in male F-344 rats. Toxicol. Sci. 96, 237–245.
Landrigan, P.J., Sonawane, B., Mattison, D., McCally, M., Garg, A., 2002. Chemical contam-inants in breast milk and their impacts on children's health: an overview. Environ.Health Perspect. 110, A313–A315.
Lankova, D., Lacina, O., Pulkrabova, J., Hajslova, J., 2013. The determination ofperfluoroalkyl substances, brominated flame retardants and their metabolites inhuman breast milk and infant formula. Talanta 117, 318–325.
Mulder, G.I., 1992. Glucuronidation and its role in regulation of biological activity of drugs.Annu. Rev. Pharmacol. Toxicol. 32, 25–49.
Nagayama, J., Taksasuga, T., Tsuji, H., 2000. Comparison between brominated flame retar-dants and dioxins or organochlorine compounds in blood levels of Japanese adults.Organohalogen Compd. 48, 27–30.
Nakagawa, Y., Suzuki, T., Ishii, H., Ogata, A., 2007. Biotransformation and cytotoxicity of abrominated flame retardant, tetrabromobisphenol A, and its analogues in rat hepato-cytes. Xenobiotica 37, 693–708.
Nakao, T., Akiyama, E., Hideki, K., Mizuno, A., Aozasa, O., Akai, Y., Ohta, S., 2015.Levels of tetrabromobisphenol A, tribromobisphenol A, dibromobisphenol A,monobromobisphenol A, and bisphenol A in Japanese breast milk. Chem. Res.Toxicol. 28, 722–728.
Nicolopoulou-Stamati, P., Hens, L., Howard, C.V. (Eds.), 2001. Endocrine Disrupters: Envi-ronmental Health and Policies. Kluwer Academic Publishers, Dordrecht; Boston.
Norris, D.O., Carr, J.A. (Eds.), 2006. Endocrine Disruption: Biological Bases for Health Ef-fects in Wildlife and Humans. Oxford University Press, Oxford.
Pellizzari, E.D., Zweidinger, R.A., Erickson (U.S. Environmental Protection Agency), 1978.Environmental Monitoring Near Industrial Sites: Brominated Chemicals. Part I. Officeof Toxic Substances, Washington, D.C.
Riu, A., Maire, A.I., Grimaldi, M., Audebert, M., Hillenweck, A., Bourguet, W., Balagur, P.,Zalko, D., 2011. Characterization of novel ligands of ERα, ERβ and PPARγ: the caseof halogenated bisphenol A and their conjugated metabolites. Toxicol. Sci. 122,372–382.
53K.-L. Ho et al. / Environment International 98 (2017) 46–53
Saegusa, Y., Fujimoto, H., Woo, G.H., Inoue, K., Takahashi, M., Mitsumori, K., Hirose, M.,Nishikawa, A., Shibutani, M., 2009. Developmental toxicity of brominated flame retar-dants, tetrabromobisphenol A and 1,2,5,6,9,10-hexabromocyclododecane, in rat off-spring after maternal exposure from mid-gestation through lactation. Reprod.Toxicol. 28, 456–467.
Schauer, U.M.D., Volkel, W., Dekant, W., 2006. Toxicokinetics of tetrabromobisphenol A inhumans and rats after oral administration. Toxicol. Sci. 91, 49–58.
Shi, T., Chen, S.J., Luo, X.J., Zhang, X.L., Tang, C.M., Luo, Y., Ma, Y.J., Wu, J.P., Peng, X.Z., Mai,B.X., 2009a. Occurrence of brominated flame retardants other than polybrominateddiphenyl ethers in environmental and biota samples from southern China.Chemosphere 74, 910–916.
Shi, Z.X., Wu, Y.N., Li, J.G., Zhao, Y.F., Feng, J.F., 2009b. Dietary exposure assessment of Chi-nese adults and nursing infants to tetrabromobisphenol-A andhexabromocyclododecanes: occurrencemeasurements in foods and humanmilk. En-viron. Sci. Technol. 43, 4314–4319.
Shi, Z., Jiao, Y., Hu, Y., Sun, Z., Zhou, X., Feng, J., Li, J., Wu, Y., 2013a. Levels oftetrabromobisphenol A, hexabromocyclododecanes and polybrominated diphenylethers in human milk from the general population in Beijing, China. Sci. Total. Envi-ron. 452–453, 10–18.
Shi, Z., Wang, Y., Niu, P., Wang, J., Sun, J., Zhang, S., Wu, Y., 2013b. Concurrent extraction,clean-up, and analysis of polybrominated diphenyl ethers, hexabromocyclododecaneisomers, and tetrabromobisphenol A in human milk and serum. J. Sep. Sci. 36,3402–3410.
Szymańska, J.A., Sapota, A., Frydrych, B., 2001. The disposition and metabolism oftetrabromobisphenol-A after a single i.p. dose in the rat. Chemosphere 45, 693–700.
Tan, X.X., Luo, X.J., Zheng, X.B., Li, Z.R., Sun, R.X., Mai, B.X., 2016. Distribution of organo-phosphorus flame retardants in sediments from the Pearl River Delta in SouthChina. Sci. Total Environ. 544, 77–84.
Tang, B., Zeng, Y.H., Luo, X.J., Zheng, X.B., Mai, B.X., 2015. Bioaccumulative characteristicsof tetrabromobisphenol A and hexabromocyclododecanes in multi-tissues of preyand predator fish from an e-waste site, South China. Environ. Sci. Pollut. Res. 22,12011–12017.
Thomsen, C., Lundanes, E., Becher, G., 2001. Brominated flame retardants in plasma sam-ples from three different occupational groups in Norway. J. Environ. Monit. 23,366–370.
Thomsen, C., Lundanes, E., Becher, G., 2002. Brominated flame retardants in archivedserum samples from Norway: a study on temporal trends and the role of age. Envi-ron. Sci. Technol. 36, 1414–1418.
van der Ven, L.T.M., de Kuil, T.V., Verhoef, A., Verwer, C.M., Lilienthal, H., Leonards, P.E.G.,Schauer, U.M.D., Cantón, R.F., Liten, S., de Jong, F.H., Visser, T.J., Dekant, W., Stern, N.,Håkansson, H., Slob, W., van den Berg, M., Vos, J.G., Piersma, A.H., 2008. Endocrine ef-fects of tetrabromobisphenol-A (TBBPA) in Wistar rats as tested in a one-generationreproduction study and a subacute toxicity study. Toxicology 245, 76–89.
Wang, J., Liu, L., Pan, B., Fu, X., Zhang, G., Zhang, L., Lin, K., 2015. Distribution of metals andbrominated flame retardants (BFRs) in sediments, soils and plants from an informale-waste dismantling site, South China. Environ. Sci. Pollut. Res. 22, 1020–1033.
Zalko, D., Prouillac, C., Riu, A., Perdu, E., Dolo, L., Jouanin, I., Canlet, C., Debrauwer, L.,Cravedi, J.-P., 2006. Biotransformation of the flame retardant tetrabromo-bisphenolA by human and rat sub-cellular liver fractions. Chemosphere 64, 318–327.