Subscriber access provided by UNIV OF SOUTHERN INDIANA
Ecotoxicology and Human Environmental Health
Relative Effect Potency Estimates for Dioxin-like Compounds in Pregnant Women with Gestational Diabetes Mellitus and Blood Glucose Outcomes Based on a Nested Case-control Study Xin Liu, Lei Zhang, Jingguang Li, Jun Wang, Guimin Meng, Min Chi, Yunfeng Zhao, and Yongning Wu Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b00988 • Publication Date (Web): 31 May 2019 Downloaded from http://pubs.acs.org on June 3, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 36
Environmental Science & Technology
1
Relative Effect Potency Estimates for Dioxin-like Compounds in Pregnant
2
Women with Gestational Diabetes Mellitus and Blood Glucose Outcomes
3
Based on a Nested Case-control Study
4 5
Xin Liua,b, Lei Zhanga, Jingguang Lia,b,*, Jun Wangc, Guimin Mengd, Min Chia,e, Yunfeng Zhaoa,
6
Yongning Wu a,b
7 8
a NHC
9 10
Key Laboratory of Food Safety Risk Assessment, China National Center for Food Safety
Risk Assessment, Beijing, China b
11
State Key Laboratory of Food Science and Technology, Nanchang University, Nanchang, China
12
c
13
d Beijing
14
e
Shenzhen Center for Chronic Disease Control, Shenzhen, China Fengtai Hospital obstetrics and gynecology, Beijing, China
Taiyuan Center for Disease Control and Prevention, Taiyuan, China
15 16
Corresponding author
17
Jingguang Li (Email:
[email protected]; Telephone: +086-10-87720035; Fax: +086-10-
18
52165485). Postal address: Room 205, No.7, Panjiayuan Nanli, Chaoyang District, Beijing
19
100021, China.
20 21
1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 36
22
Abstract
23
To improve the applicability of the toxic equivalents principle for human health risk assessment,
24
systemic relative effect potencies (REPs) for dioxin-like compounds (DLCs) deriving from
25
human in vivo data are required. A prospective nested case-control study was performed to
26
determine REPs from the human serum concentration of DLCs using gestational diabetes
27
mellitus (GDM) and fasting blood glucose (FBG) as the endpoints of concern.
28
concentration of 29 DLCs from 77 cases and 154 controls were measured.
29
regression were used to estimate the effects of individual congeners on GDM and FBG,
30
respectively.
31
coefficients, βi (DLCs)/βTCDD.
32
and FBG-REP, were established and largely agree with REPs from other human studies.
33
These human-serum REPs show much smaller variation compared to the 4 to 5 orders of
34
magnitude span in REPs database for the present WHO-TEF determination.
35
established REPs fitted well with WHO-TEFs, especially for polychlorinated dibenzo-p-
36
dioxins, furans.
37
could be used to improve health risk assessment of human body burden of DLCs.
38
Keywords: Dioxin-like compounds (DLCs); Relative effect potency (REP); Toxic equivalency
39
factor (TEF); Gestational Diabetes Mellitus (GDM); Fasting blood glucose (FBG)
Serum
Logistic and linear
The REPs based on GDM and FBG were calculated from the ratios of regression Two sets of consistent human serum-based REPs, i.e., GDM-REP
Moreover, the
These REPs reflecting real human exposure scenarios exhibited validity and
40
2 ACS Paragon Plus Environment
Page 3 of 36
41 42 43
Environmental Science & Technology
TOC art
44
3 ACS Paragon Plus Environment
Environmental Science & Technology
45
Page 4 of 36
1. Introduction
46
Polychlorinated dibenzo-p-dioxins, furans (PCDD/Fs) and dioxin-like polychlorinated
47
biphenyls (DL-PCBs) are a class of structurally related chemicals and commonly occur in the
48
environmental media and food products.1
49
distance transport and bioaccumulation, this notorious class of chemicals continuously generate
50
environmental health issues that capture public attention.
Due to environmental degradation-resistance, long-
51
Evaluating the risks for human health that are associated with PCDD/Fs and DL-PCBs
52
exposure has led to a proposal of using toxic equivalents (TEQs) methodology, in which each
53
compound has been assigned a toxic equivalency factor (TEF) by the World Health
54
Organization (WHO) that reflects its potency to exert an aryl hydrocarbon receptor (AhR)
55
involved biological or toxicological effect compared with the most potent congener, i.e.,
56
2,3,7,8-TCDD.2
57
effect potency (REP) database derived from in vivo rodent experiments and supported by in
58
vitro data.3
59
doses rather than concentrations measured in the blood or tissues (internal dose); however, this
60
has evoked wide concerns on the use of such “intake” TEFs for TEQ determinations in the
61
human body burden of dioxin-like compounds (DLCs).4
62
REP values may also be an important issue to consider.5
63
showed that various types of human cell models, both primary cells and carcinoma cell lines,
64
were much less sensitive to PCB126 than its assigned TEF value of 0.1.
65
expert panel at a recent WHO TEF meeting specified the need to assess whether systemic, as
66
well as human-specific, TEF values would lead to a more accurate human health risk
The present TEFs estimated by the expert meeting of WHO used a relative
It is important to realize that REPs from animal studies are based on oral intake
In addition, species differences in
4 ACS Paragon Plus Environment
For example, previous studies
6, 7
Therefore, an
Page 5 of 36
Environmental Science & Technology
67
assessment for DLCs.3
68
project SYSTEQ was initiated, with the main objective of developing a set of in vivo human
69
systemic REPs.
70
from thyroid volume (TV), serum free thyroxine (FT4) concentrations, cytochrome P450 (CYP)
71
1A1 and CYP1B1 mRNA expression endpoints were established from this project.8
72
Additionally, considering the joint toxicological trend in real-life mixed exposure scenarios,
73
Larsson et al.9
74
DLCs at the same time and involved multiple AhR-mediated responses using multivariate
75
statistical analysis, which was also carried out under the EU SYSTEQ project.
76
developed CTF was comparable with the TEF scale and regarded as more valid than studying
77
each DLC congener one by one. Another two in vivo rodent models derived plasma- or tissue-
78
based REPs in EU SYSTEQ project indicated that the systemic REP values of a number of
79
DLC congeners may substantially deviate from the present WHO-TEF values.10,
80
Consequently, a greater number of human studies, involving a variety of health outcomes, are
81
warranted to promote estimating more reliable TEF for human health risk assessment.
In this context, the European Union (EU) 7th Framework Programme
Presently, two studies with four sets of human blood-based REPs derived
developed a consensus toxicity factor (CTF) approach that investigated all
The newly
11
82
In the present study, REPs for the systemic human serum concentration of DLC
83
components were determined in a population of pregnant women using gestational diabetes
84
mellitus (GDM) and fasting blood glucose (FBG) concentration as the outcomes of interest.
85
total of 17 PCDD/Fs and 12 DL-PCBs with halogen substituents on the 2,3,7,8-positions were
86
prospectively detected prior to GDM and FBG measurement to investigate the causal
87
relationship between DLC exposure and outcomes using a nested case-control study.
88
Thereafter, we used a regression-based approach that compared the outcome/concentration
5 ACS Paragon Plus Environment
A
Environmental Science & Technology
Page 6 of 36
89
regression coefficients of the individual DLC congeners with the outcome/concentration
90
regression coefficient of the TCDD index to calculate REPs.
91
2. Materials and Methods
92
2.1 Study design and population
93
The participants of our study were recruited at Maternal and Child Health Hospital in
94
Xicheng district of Beijing, China.
95
control study which initialed from August 2013 to June 2015.
96
enrolled in this study during their initial antenatal care visit in the study hospital.
97
were eligible if they were free of pre-diabetes and a family history of diabetes.
98
duration of the study, a total of 439 pregnant women were recruited and agreed to donate a
99
blood sample between 9 and 13 weeks of gestation.
All pregnant women were informed of
100
the purpose of the study and provided signed consent.
The whole procedure of study had the
101
approval from the Ethics Committee of China National Center for Food Safety Risk Assessment.
102
All eligible pregnant women in the cohort were routinely screened for GDM at 24–28 weeks of
103
gestation.
104
GDM across the entire study population.
105
from the cohort as paired controls (mathcad by age: ± 2 years).
106
Our research was designed as a prospective nested caseHealthy pregnant women were Participants For the
At the end of the study period, a total of 77 pregnant women were diagnosed with For each case, two healthy women were selected
Maternal age was related to DLC exposure and GDM risk and was identified as a strong
107
confounding variable; consequently, this was used as a pair-matched factor.
108
the other potential confounders such as body mass index (BMI), cigarette and alcohol
109
consumption, parity, occupational exposure, and a family history of diabetes were obtained
110
from medical records or questionnaires.
Information for
All of the study pregnant women were found to be
6 ACS Paragon Plus Environment
Page 7 of 36
Environmental Science & Technology
111
primiparas with no cigarette smoking and alcohol consumption during pregnancy, and were
112
free from occupational exposure.
113
as pregnancy body weight (kg) divided by height-squared (m2).
114
2.2 Blood collection and chemical analysis
115
The maternal BMI was calculated at the first prenatal visit
Fasting venous blood samples were collected from the participants by nurses during their
116
first trimester.
117
segregated and preserved at -40oC until further analysis.
118
and 154 pair-matched controls were selected to measure 17 PCDD/Fs and 12 DL-PCBs, as
119
designated by the WHO.
The method for measuring these DLC congeners in serum samples
120
is described elsewhere.12
In brief, each thawed serum sample (about 2 mL) was spiked with
121
13C-labeled
122
Ontario, Canada) and then mixed with 15 g of diatomite.
123
extracted by Accelerated Solvent Extraction (ASE) 350 (ThermoScientific, Sunnyvale,
124
California, USA).
125
and carbon column (CAPE Technologies, South Portland, Maine, USA).
126
were then concentrated to nearly dryness and then resuspended in 20 μL nonane.
127
recovery standards were spiked prior to instrumental analysis.
128
measured by a high-resolution gas chromatography/high resolution mass spectrometry
129
(ThermoFisher Scientific, DFSTM Magnetic sector, Bremen, Germany).
130
then separated on a 60m DB-5 MS capillary column (0.25 mm id, 0.25 μm df;
131
Agilent Technologies, Palo Alto, California, USA).
132
were performed using U.S. EPA isotope dilution methods 1613 and 1668.13,
The whole blood sample was centrifuged immediately, and the serum was Serum samples from 77 GDM cases
internal standards (EPA1613l-LCS and P48-W-ES, Wellington Laboratories Inc., The analytes in mixture were
The extracts were subsequently purified using an acid silica gel column The elution fractions 13C-labeled
PCDD/Fs and DL-PCBs were
The analytes were
Qualitative and quantitative analyses
7 ACS Paragon Plus Environment
14
Total
Environmental Science & Technology
Page 8 of 36
133
cholesterol (CHO) and triglycerides (TG) in serum were measured by an automatic
134
biochemistry analyzer (TBA-120FR, Toshiba, Tokyo, Japan) to estimate the total serum lipids
135
by the following formula: Total lipids (mg/dL) = 2.2×cholesterol + triglycerides + 62.3.
136
Lipid-adjusted concentrations of DLC congeners were calculated by diving serum DLCs
137
concentrations by total lipid concentration.
138
15
All samples were analyzed by the laboratory analyst who was blinded to the case-control
139
status.
140
standard reference material (SRM-1957) from the National Institute of Standards and
141
Technology (NIST).
142
system contamination and as a quality control to assess laboratory precision during the entire
143
analytical process.
144
certified reference ranges provided by the manufacturer (Supplementary Table S1).
145
recovery of
146
PCDD/Fs, and from 58 % to 92 % for DL-PCBs, thus meeting the requirements of U.S. EPA
147
1613 and 1668.
148
2.3 Assessment of blood glucose and GDM
149
Each analytical batch consisted of 10 serum samples, one procedure blank and one
Procedure blank and SRM-1957 were analyzed to identify potential
The measured values of each analyte in SRM-1957 samples fell in the
13C-labled
The
internal standards across all samples ranged from 36 % to 108 % for
A “One-Step” approach, involving a 75-g oral glucose tolerance test (OGTT) without prior
150
plasma glucose screening was used to screen for GDM.
151
International Association of Diabetes and Pregnancy Study Groups, and has been adopted in
152
China.16
153
The fasting venous blood sample was collected before 9 a.m. to analyze the fasting glucose
154
concentration.
This strategy is established by the
Pregnant women were tested in the morning after overnight fasting for at least 8 h.
Then, blood samples were taken at intervals of 1 h and 2 h after drinking a
8 ACS Paragon Plus Environment
Page 9 of 36
Environmental Science & Technology
155
75-g glucose load to measure the postprandial glucose concentrations.
156
if one or more of the following threshold values were met or were exceeded: 5.1 mmol/L
157
(fasting glucose), 10.0 mmol/L (1 h glucose), and/or 8.5 mmol/L (2 h glucose).
158
2.4 Statistical analysis
159
GDM was diagnosed
Total TEQs were estimated based on multiplying the individual concentrations of
160
PCDD/Fs and DL-PCBs by their respective TEF value, as proposed by the WHO.
161
potencies of each individual congener were estimated based on comparing the magnitude of the
162
regression coefficient (β) for adverse outcomes, which underwent regression analysis with the
163
serum lipid-adjusted mass concentration of the individual DLC congeners.17, 18 Here, we used
164
conditional logistic regression and general linear regression model to estimate the effects of
165
individual DLC congeners on GDM risk and blood glucose, respectively.
166
regression with backward elimination (p > 0.10) was used to query the selection of 2,3,7,8-
167
TCDD as the index (reference) compound in concurrence with other DLCs.
168
regression analysis with all DLC congeners, the cross-tabulation for samples > the maximum
169
limit of detection (MDL) reduced the number of individuals to an insufficient number to
170
construct a regression model; thus, values below the MDL were set to zero in final analysis.
171
The regression analysis of the present study was ultimately conditioned by maternal age and
172
serum lipids, and adjusted only for pregnancy BMI.
173
and DL-PCBs congeners were calculated as the ratio of the β coefficient obtained for the ith
174
congener to β coefficient for 2,3,7,8-TCDD: βi/βTCDD.
175
blood glucose endpoints were compared with human blood-based REPs derived by other
176
investigators,17,
18
The relative
The multiple
In multiple
The REPs of the individual PCDD/Fs
The REPs derived from GDM and
WHO-TEF values,3 WHO-consensus toxicity factors (CTF),9 and with
9 ACS Paragon Plus Environment
Environmental Science & Technology
177 178
Page 10 of 36
published REP values from an in vivo database 19 using regression or correlation analysis. SPSS Statistical software version 24 (IBM Co., Armonk, NY, USA) and R software
179
version 3.5.2 (R Core Team) were used for the calculation and data visualization.
180
tests were two-tailed, and the statistically significant was set at p < 0.05.
181
3. Results
182
3.1 Participant characteristics
Significance
183
The analyzed population consisted of 77 GDM cases (mean age ± SD: 28.9 ± 4.6 years)
184
and 154 pair-matched controls (28.9 ± 2.8 years); the overall mean age (± SD) of the study
185
population was 28.9 ± 3.8 years.
186
this difference was not statistically significant (22.4 ± 3.0 vs. 21.7 ± 2.8, p =0.11).
187
and control women, the mean (± SD) of fasting blood glucose was 4.8 ± 0.7 mmol/L and 4.4 ±
188
0.4 mmol/L, respectively.
189
were 9.9 (7.3–13.2) and 6.9 (5.8–9.2) pg/g lipids, respectively.
190
rate and median serum concentrations of 29 individual PCDD/Fs and DL-PCBs congeners are
191
shown in Supplementary Table S2.
192
3.2 Index compound identification and REPs for DLC congeners
193
GDM cases showed a slightly higher BMI than controls but For GDM
The median (interquartile range) of total TEQ for GDM and control Data relating to the detection
There is general agreement that the index congener for DLCs should be potent with regard
194
to the expected endpoints.20
195
to the incidence of GDM, 2,3,7,8-TCDD was the most potent congener (Table S3).
196
regression analysis, with FBG as the endpoint of interest, 2,3,7,8-TCDD was also retained in
197
the multiple regression model that showed the positive association.
198
regression analysis did not confirm the potent role for 2,3,7,8-TCDD when 1 h and 2 h
Conditional logistic regression analysis showed that with respect
10 ACS Paragon Plus Environment
In linear
However, multiple linear
Page 11 of 36
Environmental Science & Technology
199
postprandial blood glucose was used as the dependent variables (endpoints).
200
coefficient (β), and the REPs for DLC congeners, as derived from GDM and FBG outcomes
201
are presented in Supplementary Table S4.
The regression
202
The regression β of GDM against DLC congener levels were positive for all individuals
203
of the exposure mixture except for 1,2,3,4,7,8-HxCDD, OCDD, 1,2,3,4,7,8-HxCDF,
204
2,3,4,6,7,8-HxCDF, 1,2,3,7,8,9-HxCDF, OCDF and for DL-PCB congeners 105, 123 and 189.
205
This means that increased exposure to these congeners, including 2,3,7,8-TCDD, was positively
206
associated with GDM risk.
207
congeners were positively associated with FBG concentration.
208
1,2,3,7,8-PeCDD, 1,2,3,4,7,8-HxCDD, 2,3,4,6,7,8-HxCDF, 1,2,3,7,8,9-HxCDF, OCDF and
209
PCB105 exhibited the opposite association with FBG in comparison to 2,3,7,8-TCDD.
210
order to comply with the TEF methodology assumption that DLCs have a common mode of
211
action, REPs were calculated only for those congeners acting in the same direction as the index
212
compound.
213
human studies
214
allowing easy comparison.
215
3.3 Comparison of REPs calculated from various endpoints in previous human studies
In terms of FBG endpoints, 3 PCDD, 7 PCDF and 11 DL-PCB
All REPs from this study are listed in Table 1. 17, 18
However, the congeners of
In
In addition, REPs from other
are also given in Table 1, along with WHO TEFs2005 and CTFs,9 thus
216
REPs originating from GDM and FBG outcomes were compared with REPs derived from
217
previous human studies involving thyroid volume, serum FT4, CYP1A1 and CYP1B1
218
expression outcomes.
219
in Figure 1.
220
level (asterisk) between row variable and column variable. Within our own study, the
Correlations linking REPs derived from these outcomes are presented
This correlation matrix shows correlation coefficients (number) and significant
11 ACS Paragon Plus Environment
Environmental Science & Technology
Page 12 of 36
221
conditional logistic regression β coefficient derived GDM-REPs were significantly correlated
222
with linear regression β coefficient derived FBG-REPs (Pearson r = 0.930, p < 0.001).
223
Moreover, GDM-REPs were also significantly correlated with REPs derived from CYP 1A1 (r
224
= 0.898, p < 0.001), thyroid volume (r = 0.744, p = 0.014) and FT4 (r = 0.900, p < 0.001).
225
With regards to FBG-REPs, a significant correlation was observed with CYP 1A1-REPs (r =
226
0.763, p = 0.01) and FT4-REPs (r = 0.777, p = 0.02).
227
TV-REPs, we found that r = 0.617 and p = 0.07, thus showing that the REPs derived from these
228
two outcomes were associated at a marginal level of significance.
229
When comparing of FBG-REPs with
Figure 2 illustrates the distribution of REP values normalized to the respective congener
230
REPs data set.
231
outcomes were relatively smaller than PCDF and PCB congeners.
232
group, CYP 1B1-REPs for 1,2,3,4,7,8-HxCDF and 1,2,3,4,6,7,8-HpCDF were above the
233
normalized median value (0.5) that far from other outcomes derived REPs.
234
was observed for the FBG-REPs of 1,2,3,4,7,8,9-HpCDF which showed a high deviation.
235
Only two REPs (originated from CYP1A1and FT4) were obtained for OCDF, which also
236
exhibited significant deviation from each other.
237
degrees of value distribution were observed, particularly for the REPs of PCB77, 105 and 189.
238
However, almost all of the human serum derived REPs were within or marginally above one
239
order of magnitude.
240
3.4 Comparison of human REPs with WHO-TEFs and CTFs
241 242
The deviation of REPs for PCDD congeners derived from six different In the PCDF congeners
A similar pattern
With regards to DL-PCBs, the high discrete
In order to further compare our REPs with respective WHO-TEFs, the TEFs were set to 1 and the deviation of REPs from TEFs were calculated as REP/TEF (Figure 3).
12 ACS Paragon Plus Environment
In the PCDD
Page 13 of 36
Environmental Science & Technology
243
congener group, GDM-REPs and FBG-REPs were comparative to respective WHO-TEFs.
244
Furthermore, the deviations were all within half an order of magnitude of their WHO-TEFs
245
except for the FBG-REP value for 1,2,3,4,6,7,8-HpCDD.
246
REPs of 2,3,7,8-TCDF, 2,3,4,7,8-PeCDF, 1,2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF derived
247
from GDM or FBG were approximately equal to their respective WHO-TEFs, or the deviations
248
were within the WHO-TEF half-log uncertainty range.
249
approximately one order of magnitude higher than their WHO-TEF values.
250
the DL-PCB congeners group, most congener REP values were much higher than their WHO-
251
TEFs, with deviations spanning one or three orders of magnitude.
252
(based on GDM) and PCB123 (based on FBG) were less than 1 order of magnitude higher than
253
their WHO-TEFs.
254
and FBG endpoints, was close to the WHO-TEF value.
255
to the WHO-TEF value, or the half-log uncertainty border line.
256
was within half an order of magnitude.
257
In the PCDF congener group, the
The REPs for 1,2,3,7,8-PeCDF were With regards to
The REPs for PCB77
It is also worth noting that the REP for PCB126, derived from both GDM The two REPs of PCB169 were close The GDM-REP of PCB118
Figure 4 depicts the combination of PCDD/Fs and DL-PCBs REP values in relation to the
258
WHO TEFs and consensus toxicity factors (CTF).
259
correlated significantly with the WHO-TEFs (GDM, Pearson r = 0.84, p < 0.001; FBG, r =
260
0.79, p < 0.001).
261
GDM-REPs and log FBG-REPs fitted well with the rat log CTF with correlation coefficients
262
of 0.81 (p < 0.001) and 0.80 (p < 0.001), respectively; however, these REPs were not
263
significantly correlated to human log CTFs.
264
above the central axis were higher than TEF or CTF values, and vice versa.
The GDM-REPs and FBG-REPs
In addition, by comparing the REPs with CTFs, we observed that both log
According to the estimates, congener potencies
13 ACS Paragon Plus Environment
DLCs were then
Environmental Science & Technology
Page 14 of 36
265
divided into PCDD/Fs group and DL-PCBs group to investigate the correlation between the
266
respective subset of REP values and either TEF or CTF.
267
congeners, i.e. PCDD/Fs and DL-PCBs, also correlated well with the TEF and rat CTF subset,
268
respectively, while FBG-REPs for PCDD/Fs did not show a notable correlation with rat CTF
269
(r = 0.561, p = 0.073).
270
3.5 Comparison of human REPs with data from a published in vivo database
Figure 5 shows that the grouped DLC
271
In order to investigate our human REPs in a broader context, we extracted published in
272
vivo experimental REPs from the REP2004 database19 and also identified newly - derived in vivo
273
REPs published between 2005 and 2018.10, 11, 17, 18, 21-24 A violin plot was used to visualize the
274
distribution of the updated in vivo REP values and their probability density (Figure 6A for
275
PCDD/Fs and Fig. 6B for DL-PCBs).
276
predominantly located between the published minimum and maximum values.
277
congener group, the REP for 1,2,3,7,8-PeCDD derived from GDM lies at the 50th percentile of
278
the REP distribution.
279
and CTF values, were close to the maximum value of the REP distribution, while the WHO
280
TEF was above the maximum value.
281
lay between the 50th and 75th percentiles, and the GDM-REP for TCDF was close to the 75th
282
percentile.
283
derived from FBG, were around the 75th percentile.
284
TEF was usually assigned a value within the 50th and 75th percentile of the REPs database
285
distribution, with a general inclination towards the 75th percentile in order to be protective for
286
human health.3
As shown in the violin plot, our REPs were In the PCDD
The 1,2,3,4,6,7,8-HpCDD REP value derived from GDM in our study,
As for the PCDF congener group, the TCDF REP values
REP values for 2,3,4,7,8-PeCDF, 1,2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF, as As indicated by the WHO expert panel,
With the addition of newly identified in vivo REP values arising from research
14 ACS Paragon Plus Environment
Page 15 of 36
Environmental Science & Technology
287
carried out between 2005 and 2018, the TEF values for OCDD, 1,2,3,4,6,7,8-HpCDF and
288
OCDF fell below the 50th percentile of the REPs distribution range.
289
Figure 6B illustrates the wide variation in REPs for different DL-PCBs, in which the REP
290
values span 4-5 orders of magnitude, depending on the congener.
291
for PCB77 lay within the 50th and 75th percentiles.
292
were greater than the maximum value of REP distribution.
293
CTF rat value and the two REP values in our study were similar to each other and were around
294
the median of the REPs distribution.
295
while the PCB123 FBG-REP value lay within the 25th and 50th percentiles.
296
167 and 189 lay within the 50th and 75th percentiles.
297
ortho PCBs were all below the 50th percentile of the REPs distribution.
298
4. Discussion
299
GDM-REP and FBG-REP
REPs (GDM- and FBG-based) for PCB81 As for PCB126, the TEF value,
REPs for PCB114 and 156 lay above the 75th percentile, REPs for PCB157,
It should be noted that the TEF for mono-
The risk assessment of human exposure to PCDD/Fs and DL-PCBs remains a continuous
300
challenge for regulatory authorities.
301
every few years in order to take into account newly derived in vivo REP values and reduce
302
uncertainties.3
303
recommends deriving systemic TEFs for PCDD/Fs and DL-PCBs for use in human
304
epidemiological studies in the most recent scientific opinion.25
305
REPs become particularly of interest in updating more appropriate and accurate TEF to improve
306
human risk assessment.
307
of PCDD, PCDF and DL-PCBs were established using GDM and FBG as endpoints of interest.
308
Compared with other REPs derived from human studies using four independent endpoints,17, 18
Expert panels from the WHO update the TEF of DLCs
The EFSA Panel on Contaminants in the Food Chain (CONTAM)
Therefore, in vivo systemic
In the present study, two sets of human serum-based systemic REPs
15 ACS Paragon Plus Environment
Environmental Science & Technology
Page 16 of 36
309
we found that the REPs derived from these various outcomes in different human populations
310
environmentally exposed to different levels of DLCs were significantly interrelated.
311
Moreover, the REP values for PCDD/Fs originating from different study populations show
312
relatively small variation compared to the 2 to 4 orders of magnitude span in the REPs database
313
from in vivo or in vitro studies.
314
high correlation, both with the WHO-TEF values, and with CTFs developed as a novel approach
315
to establish toxicity factors for the risk assessment of PCDD/Fs and DL-PCBs.
In addition, these REPs, derived from GDM and FBG, showed
316
The prerequisite for the derivation of human-specific REPs in current study was the pre-
317
identified prospective association between the outcomes, e.g., GDM and FBG), and the
318
concentrations of DLCs in serum. Prospective study characterized by the measurement of risk
319
factors or exposures before the outcome occurs could provide stronger scientific evidence than
320
retrospective studies such as cross-sectional and case-control studies. Thus, the present study
321
examined a prospective data on GDM in pregnant women exposed to a mixture of DLCs to
322
evaluate the plausible causal relationships between DLCs and GDM as well as FBG.
323
TEQ was used to perform a cumulative risk assessment of the DLC mixtures and significantly
324
increased GDM risk, with an estimate odds ratio (OR) of 2.12 (95% CI: 1.57, 2.86)
325
(Supplementary Figure S1).
326
associated with increased FBG.
327
Institute of Environmental Health Sciences (NIEHS)/National Toxicology Program (NTP)
328
workgroup also concluded that the overall evidence from human studies is sufficient for a
329
positive association of DLCs with diabetes despite the heterogeneity of the studies.26
330
Furthermore, accumulating evidence from experimental in vitro and in vivo studies have
Total
The increment in total TEQ concentration was also significantly In addition, a systemic review conducted by the National
16 ACS Paragon Plus Environment
Page 17 of 36
Environmental Science & Technology
331
converged to support the postulate that DLCs are mechanistically implicated in an increased
332
risk of the development of diabetes or its related endpoints observed in human studies.
333
a differentiated adipocytes model, TCDD was found to significantly attenuated insulin-induced
334
glucose uptake.27
335
exposure resulted in an increased basal-insulin release in both islets and β-cell model.28
336
recently, it has been observed that dioxin and dioxin-like PCBs induced activation of the AhR
337
could subsequently affect the expression of gene that related to insulin transport, glucose
338
uptake,29 glucose homeostasis,30 lipid metabolism,31, 32 and inflammation,33, 34 etc., which play
339
a pivotal role in diabetes progression.
340
dysregulation of the link between fatty liver and insulin resistance, correspondently, chemical
341
inhibition of AhR was found that lead to decreased obesity and fatty livers in C57BL/6J mice.32
342
Studies in AhR-deficient mice demonstrate enhanced insulin sensitivity and increased glucose
343
tolerance,36 findings that are consistent with an experiment that found dioxin-exposed mice to
344
have significantly reduced insulin secretion in wild-type mice but not in AhR‐null mice.37
345
cell-based AhR ligand assay conducted in human serum revealed that AhR ligand activity is
346
associated with insulin resistance and resulting T2DM.38
347
that AhR plays a physiological function in glucose metabolism, and supported postulate that
348
sustained activation of the AhR by DLCs could contribute to the development of diabetes.
349
the basis of biological link between DLC-induced AhR activation and diabetes or its related
350
endpoints as well as the identified prospective association between DLCs exposure and GDM
351
in our cohort, estimating REPs based on GDM and FBG was considered reasonable.
352
As in
In an ex-vivo mouse islets and the β-cell-line study, 2,3,7,8-TCDD More
Notably, Lu et al.35 found that AhR activation result in
A
Overall, these findings suggested
On
Most of the previously published REP values for PCDD/Fs and DL-PCBs are based on in
17 ACS Paragon Plus Environment
Environmental Science & Technology
Page 18 of 36
353
vitro or on in vivo animal models.19
354
exposure of humans is more complex; thus human-specific responses and the body burden of
355
DLC concentrations could lead to reliable potency estimates and ultimately improve human
356
risk assessments for DLCs.39
357
concentration was conducted by Trnovec et al.17 under the EU SYSTEQ project.
358
authors applied a novel approach which compared the outcome/concentration regression
359
coefficients of the individual DLC congeners with the outcome/concentration regression
360
coefficient of the TCDD index with which to calculate the REPs.
361
were subsequently complemented and corroborated by Wimmerová et al.18 using more sensitive
362
molecular biomarkers, CYP1A1 and CYP1B1, known to be able to regulated by the aromatic
363
hydrocarbon receptor (AhR).40
364
in eastern Slovakia, an area known to be contaminated by the organochlorine compounds.
365
accumulated concentration of PCDD/Fs and DL-PCBs for these participants was 23.3 pg/g
366
lipid-adjusted TEQ which was approximately 3-fold higher than our present study population
367
(7.7 pg/g lipid TEQ).
368
with background exposure mainly via the diet which could compensate for the low exposure
369
range response for DLCs.
370
FBG in our study were significantly correlated with REPs based on TV, FT4 concentration and
371
CYP1A1 expression outcomes in previous studies of the Slovakian population,17, 18 but also
372
REPs derived from these two different study populations show a relatively small variation, as
373
demonstrated by the normalized REP distribution plot in which the deviations were almost all
374
within or marginally above one order of magnitude.
Compared with laboratory exposures, the environmental
The first human in vivo analysis of REPs based on blood DLCs These
Their thyroid outcome data
These two studies were conducted in a same population living The
The participants in our study were recruited from a general population
It is worth noting that not only the REPs derived from GDM and
Moreover, the REPs for PCDD and
18 ACS Paragon Plus Environment
Page 19 of 36
Environmental Science & Technology
375
PCDF in human blood-based studies using clinical outcomes (GDM, FBG, TV and FT4), or
376
molecular biomarker endpoints (CYP1A1 and CYP1B1), are in good accordance with their
377
assigned WHO-TEF values and the deviations mostly fall within the half-log uncertainty.
378
concordant behavior between REPs derived from human studies and the WHO TEFs imply that
379
the point estimation for TEF value by the WHO expert panel, based on the distribution of REPs
380
between 50th and 75th percentile, is reasonable for human health risk assessment for PCDD and
381
PCDF.
382
concentration of TCDD, 1-PeCDD and 4-PeCDF with an extra-hepatic response (hepatic
383
ethoxyresorufin-O-deethylase activity, and hepatic CYP 1A1, 1A2 and 1B1 expression) were
384
also suggested to provide a more accurate prediction of REPs for humans under environmental
385
exposure conditions.11, 41
386
The
Within the EU SYSTEQ project, systemic REPs based on rat or mice blood
In the current study, the REPs for mono-ortho PCB lie within the published extremes,
387
whereas most of them were seriously deviated from their WHO-TEFs.
388
this might be that mono-ortho PCBs elicit a diverse spectrum of non-AhR-mediated pathways
389
in order to exert diabetogenic or hyperglycemic effects.
390
for PCB169 and 118, based on GDM outcome, were within the +/- half log uncertainty of the
391
WHO-TEF values.
392
Slovakian population-based study, also fell within the half log uncertainty.
393
CYP1B1-REPs for PCB105, 114, 123, 156, 157 and 189 in the previous Slovakian population
394
studies were approximately 1–10 times higher than WHO-TEF values.
395
and FBG-derived REPs for these mono-ortho PCBs were about 10–1000 times greater than
396
WHO-TEF.
One explanation for
It is worth highlighting that the REPs
Consistently, the REPs of these two congeners, based on CYP1B1 in However,
Moreover, our GDM-
The REPs for mono-ortho PCBs were not obtained in Slovakian population based
19 ACS Paragon Plus Environment
Environmental Science & Technology
Page 20 of 36
397
on thyroid volume, FT4 concentration and CYP1A1 expression outcomes due to opposite
398
responses against index compound.
399
from human studies with different endpoints for mono-ortho PCBs.
400
commonly be observed in REPs for the same congener and for similar outcomes in different
401
species or for the same species with different outcomes.
402
REPs within in vivo and in vitro database spans several orders of magnitude.19
403
variation inherent in the REPs database for mono-ortho PCBs has aroused serious concern with
404
regards to the expert WHO TEF panel meeting in 2005.
405
the time, the WHO expert panel expressed low confidence in greater REP values for TEF
406
assignment of certain mono-ortho PCBs.
407
50th percentile of the REP distribution was used to uniformly set the TEFs for all mono-ortho
408
PCBs at 0.00003 in the 2005 WHO TEF expert meeting.
409
data to set scientifically balanced TEF values hinders accurate human risk assessment for mono-
410
ortho PCBs.
411
debate in recent years, because several studies using different human cell models, proved that
412
PCB126 was much less potent than in vivo rodent studies.42-45
413
cell-based assays for PCB126 is 0.0033, which is approximately 30-fold lower than its assigned
414
WHO-TEF.6
415
a REP of 0.017 with CYP1B1 as an endpoint.
416
to the margin-bottom of the half-log uncertainty for the TEF value (0.03).
417
low sensitivity, in terms of human exposure linking to outcomes, and significant variation for
418
PCB126 were also not observed as the REPs for PCB126 were 0.152 (GDM-based) and 0.0745
This scenario reflects the large variation REPs derived Great variation can
The distribution of mono-ortho PCBs The significant
Based upon information available at
Due to the lack of sufficient experimental data, the
Obviously, the lack of sufficient
In terms of non-ortho DL-PCBs, the REP of PCB126 has been the focus of much
The median REP from human
As for the human blood-based REPs of PCB126, Wimmerová et al.18 obtained This human-specific REP for PCB126 is close
20 ACS Paragon Plus Environment
In the present study,
Page 21 of 36
Environmental Science & Technology
419
(FBG-based); these are comparable to its TEF value of 0.1 and the deviations were within +/-
420
half log uncertainty.
421
human risk assessment as toxicokinetic absorption, distribution, metabolism and excretion
422
could be the main issues of concern.6, 41
423
derived systemic TEFs.
424
human-specific PCB126 REP values.
425
are now required to derive appropriate and accurate TEFs for DL-PCBs in human risk
426
assessment.
Human-relevant in vitro cell models may not be best suited for reliable
Human risk assessment may better rely on human-
Consequently, more research and understanding are needed on In particular, more human blood-based investigations
427
One of the limitations of this study is the limited sample size of our study population.
428
Future studies, using a larger population with different exposure ranges and linking to various
429
adverse responses, are now needed to validate concordant behavior in the estimation of human-
430
specific REPs.
431
although several potential confounders were controlled in our study, we cannot exclude other
432
important issues inherent in human studies such as mixture exposure effects, residual
433
confounding factors, the effect of unmeasured AhR-active compounds which may bias the
434
effect estimation with studied outcomes.
435
derived REPs in present study underwent sufficient comparison with established authority TEF,
436
CTF and REP values, and suggested concordance.
437
It is well-known that humans are exposed to a diverse of environmental factors,
While the DLCs exposure and adverse outcome
In conclusion, the estimation of human serum-based REPs for DLCs was reliable and
438
considered to be of high relevance for human risk assessment.
439
values intend to apply weighting factors for different types of studies to the existing REP data
440
set.
Future updates of the TEF
In this context, human systemic REPs are valuable and may be considered to provide a
21 ACS Paragon Plus Environment
Environmental Science & Technology
441
high weighting in order to derive more accurate TEF values for human body burden toxic
442
equivalents (TEQ) determinations. Although the present human data supports the TEF values
443
for PCB 126 and 169, more studies of the REPs for mono-ortho PCBs are warranted due to the
444
inherent significant variation. Additionally, the future research should also focus on what PCB
445
chemicals are actually exposed in people.
446
Acknowledgements
447
The authors gratefully acknowledge all the mothers who collaborated with this study and
448
donated serum samples.
449
Foundation of China [grant numbers 21537001, 21477030 and 21507018], The National Key
450
Research and Development Program of China [grant number 2017YFC1600500], and the
451
Strategic Priority Research Program of the Chinese Academy of Sciences [grant number XDB
452
14010100].
453
Supporting Information Available
454
Additional information including supplemental tables (Tables S1-S4) and supplemental figure
455
(Figure S1). This information is available free of charge via the internet at http://pubs.acs.org.
456
Disclosures
457
The authors declare no competing financial interest.
This research was supported by the National Natural Science
458
22 ACS Paragon Plus Environment
Page 22 of 36
Page 23 of 36
Environmental Science & Technology
459 460
References:
461
(1) Malisch, R.; Kotz, A., Dioxins and PCBs in feed and food-review from European
462
perspective. Sci. Total Environ. 2014, 491, 2-10.
463
(2) Van den Berg, M.; Birnbaum, L.; Bosveld, A. T.; Brunstrom, B.; Cook, P.; Feeley, M.;
464
Giesy, J. P.; Hanberg, A.; Hasegawa, R.; Kennedy, S. W.; Kubiak, T.; Larsen, J. C.; van
465
Leeuwen, F. X.; Liem, A. K.; Nolt, C.; Peterson, R. E.; Poellinger, L.; Safe, S.; Schrenk, D.;
466
Tillitt, D.; Tysklind, M.; Younes, M.; Waern, F.; Zacharewski, T., Toxic equivalency factors
467
(TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 1998,
468
106, (12), 775-92.
469
(3) Van den Berg, M.; Birnbaum, L. S.; Denison, M.; De Vito, M.; Farland, W.; Feeley, M.;
470
Fiedler, H.; Hakansson, H.; Hanberg, A.; Haws, L., The 2005 World Health Organization re-
471
evaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like
472
compounds. Toxicol. Sci. 2006, 93, (2), 223-241.
473
(4) van Ede, K. I.; van Duursen, M. B.; van den Berg, M., Evaluation of relative effect potencies
474
(REPs) for dioxin-like compounds to derive systemic or human-specific TEFs to improve
475
human risk assessment. Arch. Toxicol. 2016, 90, (6), 1293-1305.
476
(5) USEPA (U.S. Environmental Protection Agency). Framework for application of the toxicity
477
equivalence methodology for polychlorinated dioxins, furans and biphenyls in ecological risk
478
assessment; EPA/100/R-08/004; Office of the Science Advisor and Risk Assessment Forum:
479
Washington, DC, 2008; www.epa.gov/sites/production/files/2013-09/documents/tefs-draft-
480
052808-0804.pdf.
481
(6) van Duursen, M. B. M.; van Ede, K. I.; van den Berg, M., One TEF concept does not fit all:
482
The case for human risk assessment of polychlorinated biphenyls. Curr Opinion Toxicol 2017,
483
2, 103-108.
484
(7) Farmahin, R.; Jones, S. P.; Crump, D.; Hahn, M. E.; Giesy, J. P.; Zwiernik, M. J.; Bursian,
485
S. J.; Kennedy, S. W., Species-specific relative AHR1 binding affinities of 2,3,4,7,8-
486
pentachlorodibenzofuran explain avian species differences in its relative potency. Com.
487
Biochem. Phys. C. 2014, 161, 21-25.
488
(8) European Commission Website; https://cordis.europa.eu/project/rcn/90542/reporting/en. 23 ACS Paragon Plus Environment
Environmental Science & Technology
489
[Last update: 21 Jun 2017].
490
(9) Larsson, M.; van den Berg, M.; Brenerová, P.; van Duursen, M. B. M.; van Ede, K. I.; Lohr,
491
C.; Luecke-Johansson, S.; Machala, M.; Neser, S.; Pěnčíková, K.; Poellinger, L.; Schrenk, D.;
492
Strapáčová, S.; Vondráček, J.; Andersson, P. L., Consensus Toxicity Factors for
493
Polychlorinated Dibenzo-p-dioxins, Dibenzofurans, and Biphenyls Combiningin Silico Models
494
and Extensivein Vitro Screening of AhR-Mediated Effects in Human and Rodent Cells. Chem.
495
Res. Toxicol. 2015, 28, (4), 641-650.
496
(10) van Ede, K. I.; Andersson, P. L.; Gaisch, K. P. J.; van den Berg, M.; van Duursen, M. B.
497
M., Comparison of intake and systemic relative effect potencies of dioxin-like compounds in
498
female rats after a single oral dose. Arch. Toxicol. 2014, 88, (3), 637-646.
499
(11) van Ede, K. I.; Andersson, P. L.; Gaisch, K. P. J.; van den Berg, M.; van Duursen, M. B.
500
M., Comparison of Intake and Systemic Relative Effect Potencies of Dioxin-like Compounds
501
in Female Mice after a Single Oral Dose. Environ. Health Perspect. 2013, 121, (7), 847-853.
502
(12) Zhang, L.; Zhong, Y.; Liu, X.; Bao, Y.; Zhao, Y.; Wu, Y.; Cai, Z.; Li, J., Determination of
503
polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans, and dioxin-like
504
polychlorinated biphenyls in human serum using programmable-temperature vaporization gas
505
chromatography with high-resolution mass spectrometry. J. Sep. Sci. 2017, 40, (17), 3453-3461.
506
(13) USEPA (U.S. Environmental Protection Agency). Method 1613: Tetra-through Octa-
507
Chlorinated Dioxins and Furans by Isotope Dilution HRGC/HRMS; EPA 621-B-94-005;
508
United States Environmental Protection Agency Office of Water: Washington, DC, 1994;
509
www.well-labs.com/docs/epa_method_1613b_1994.pdf.
510
(14) USEPA (U.S. Environmental Protection Agency). Method 1668, Revision A: Chlorinated
511
biphenyl congeners in water, soil, sediment, and tissue by HRGC/HRMS; EPA 821-R-00-002;
512
United States Environmental Protection Agency Office of Water: Washington, DC, 1999;
513
www.o2si.com/docs/epa-method-1668A.pdf.
514
(15) Phillips, D. L.; Pirkle, J. L.; Burse, V. W.; Bernert, J. T.; Henderson, L. O.; Needham, L.
515
L., Chlorinated hydrocarbon levels in human serum: effects of fasting and feeding. Arch.
516
Environ. Con. Tox. 1989, 18, (4), 495-500.
517
(16) Yang, H. X., Diagnostic criteria for gestational diabetes mellitus (WS 331-2011). Chin
518
Med J (Engl) 2012, 125, (7), 1212-3. 24 ACS Paragon Plus Environment
Page 24 of 36
Page 25 of 36
Environmental Science & Technology
519
(17) Trnovec, T.; Jusko, T. A.; Šovcíková, E.; Lancz, K.; Chovancová, J.; Patayová, H.;
520
Palkovicová, L. U.; Drobná, B.; Langer, P.; Van den Berg, M.; Dedik, L.; Wimmerová, S.,
521
Relative effect potency estimates of dioxin-like activity for dioxins, furans, and dioxin-like
522
PCBs in adults based on two thyroid outcomes. Environ. Health Perspect. 2013, 121, (8), 886-
523
892.
524
(18) Wimmerová, S.; van den Berg, M.; Chovancová, J.; Patayová, H.; Jusko, T. A.; van
525
Duursen, M. B. M.; Palkovičová Murínová, Ľ.; Canton, R. F.; van Ede, K. I.; Trnovec, T.,
526
Relative effect potency estimates of dioxin-like activity for dioxins, furans, and dioxin-like
527
PCBs in adults based on cytochrome P450 1A1 and 1B1 gene expression in blood. Environ.
528
Int. 2016, 96, 24-33.
529
(19) Haws, L. C.; Su, S. H.; Harris, M.; DeVito, M. J.; Walker, N. J.; Farland, W. H.; Finley,
530
B.; Birnbaum, L. S., Development of a Refined Database of Mammalian Relative Potency
531
Estimates for Dioxin-like Compounds. Toxicol. Sci. 2006, 89, (1), 4-30.
532
(20) USEPA (U.S. Environmental Protection Agency). Supplementary guidance for conducting
533
health risk assessment of chemical mixtures; EPA/630/R-00/002; United States Environmental
534
Protection Agency Risk Assessment Forum Technical Panel: Washington, DC, 2000;
535
https://ofmpub.epa.gov/eims/eimscomm.getfile?p_download_id=4486.
536
(21) Gray, M. N.; Aylward, L. L.; Keenan, R. E., Relative Cancer Potencies of Selected Dioxin-
537
Like Compounds on a Body-Burden Basis: Comparison to Current Toxic Equivalency Factors
538
(TEFs). J. Toxicol. Env. Heal. A 2006, 69, (10), 907-917.
539
(22) Kim, E.; Suda, T.; Tanabe, S.; Batoev, V. B.; Petrov, E. A.; Iwata, H., Evaluation of
540
Relative Potencies for in Vitro Transactivation of the Baikal Seal Aryl Hydrocarbon Receptor
541
by Dioxin-Like Compounds. Environ. Sci. Technol. 2011, 45, (4), 1652-1658.
542
(23) Strapáčová, S.; Brenerová, P.; Krčmář, P.; Andersson, P.; van Ede, K. I.; van Duursen, M.
543
B. M.; van den Berg, M.; Vondráček, J.; Machala, M., Relative effective potencies of dioxin-
544
like compounds in rodent and human lung cell models. Toxicology 2018, 404-405, 33-41.
545
(24) van Ede, K. I.; van Duursen, M. B. M.; van den Berg, M., Evaluation of relative effect
546
potencies (REPs) for dioxin-like compounds to derive systemic or human-specific TEFs to
547
improve human risk assessment. Arch. Toxicol. 2016, 90, (6), 1293-1305.
548
(25) Knutsen, H. K.; Alexander, J.; Barregård, L.; Bignami, M.; Brüschweiler, B.; Ceccatelli, 25 ACS Paragon Plus Environment
Environmental Science & Technology
549
S.; Cottrill, B.; Dinovi, M.; Edler, L.; Grasl Kraupp, B.; Hogstrand, C.; Nebbia, C. S.; Oswald,
550
I. P.; Petersen, A.; Rose, M.; Roudot, A. C.; Schwerdtle, T.; Vleminckx, C.; Vollmer, G.;
551
Wallace, H.; Fürst, P.; Håkansson, H.; Halldorsson, T.; Lundebye, A. K.; Pohjanvirta, R.;
552
Rylander, L.; Smith, A.; van Loveren, H.; Waalkens Berendsen, I.; Zeilmaker, M.; Binaglia,
553
M.; Gómez Ruiz, J. Á.; Horváth, Z.; Christoph, E.; Ciccolallo, L.; Ramos Bordajandi, L.;
554
Steinkellner, H.; Hoogenboom, L. R., Risk for animal and human health related to the presence
555
of dioxins and dioxin‐like PCBs in feed and food. EFSA Journal 2018, 16, (11), 5333.
556
(26) Taylor, K. W.; Novak, R. F.; Anderson, H. A.; Birnbaum, L. S.; Blystone, C.; DeVito, M.;
557
Jacobs, D.; Köhrle, J.; Lee, D.; Rylander, L.; Rignell-Hydbom, A.; Tornero-Velez, R.; Turyk,
558
M. E.; Boyles, A. L.; Thayer, K. A.; Lind, L., Evaluation of the association between persistent
559
organic pollutants (POPs) and diabetes in epidemiological studies: A National Toxicology
560
Program Workshop Review. Environ. Health Perspect. 2013, 121, (7), 774-783.
561
(27) Hsu, H.; Tsou, T.; Chao, H.; Kuo, Y.; Tsai, F.; Yeh, S., Effects of 2,3,7,8-
562
tetrachlorodibenzo-p-dioxin on adipogenic differentiation and insulin-induced glucose uptake
563
in 3T3-L1 cells. J. Hazard. Mater. 2010, 182, (1-3), 649-655.
564
(28) Lai, K. P.; Wan, H. T.; Ng, A. H.; Li, J. W.; Chan, T. F.; Wong, C. K., Transcriptomic and
565
Functional Analyses on the Effects of Dioxin on Insulin Secretion of Pancreatic Islets and β-
566
Cells. Environ. Sci. Technol. 2017, 51, (19), 11390-11400.
567
(29) Diani-Moore, S.; Ram, P.; Li, X.; Mondal, P.; Youn, D. Y.; Sauve, A. A.; Rifkind, A. B.,
568
Identification of the aryl hydrocarbon receptor target gene TiPARP as a mediator of suppression
569
of hepatic gluconeogenesis by 2,3,7,8-Tetrachlorodibenzo-p-dioxin and of nicotinamide as a
570
corrective agent for this effect. J. Biol. Chem. 2010, 285, (50), 38801-38810.
571
(30) Zhang, W.; Sargis, R. M.; Volden, P. A.; Carmean, C. M.; Sun, X. J.; Brady, M. J., PCB
572
126 and other dioxin-like PCBs specifically suppress hepatic PEPCK expression via the aryl
573
hydrocarbon receptor. PLoS One 2012, 7, (5), e37103.
574
(31) Sato, S.; Shirakawa, H.; Tomita, S.; Ohsaki, Y.; Haketa, K.; Tooi, O.; Santo, N.; Tohkin,
575
M.; Furukawa, Y.; Gonzalez, F. J., Low-dose dioxins alter gene expression related to
576
cholesterol biosynthesis, lipogenesis, and glucose metabolism through the aryl hydrocarbon
577
receptor-mediated pathway in mouse liver. Toxicol. Appl. Pharm. 2008, 229, (1), 10-19.
578
(32) Moyer, B. J.; Rojas, I. Y.; Kerley-Hamilton, J. S.; Nemani, K. V.; Trask, H. W.; Ringelberg, 26 ACS Paragon Plus Environment
Page 26 of 36
Page 27 of 36
Environmental Science & Technology
579
C. S.; Gimi, B.; Demidenko, E.; Tomlinson, C. R., Obesity and fatty liver are prevented by
580
inhibition of the aryl hydrocarbon receptor in both female and male mice. Nutr. Res. 2017, 44,
581
38-50.
582
(33) Kim, M. J.; Pelloux, V.; Guyot, E.; Tordjman, J.; Bui, L. C.; Chevallier, A.; Forest, C.;
583
Benelli, C.; Clement, K.; Barouki, R., Inflammatory pathway genes belong to major targets of
584
persistent organic pollutants in adipose cells. Environ. Health Perspect. 2012, 120, (4), 508-14.
585
(34) Vogel, C. F. A.; Khan, E. M.; Leung, P. S. C.; Gershwin, M. E.; Chang, W. L. W.; Wu,
586
D.; Haarmann-Stemmann, T.; Hoffmann, A.; Denison, M. S., Cross-talk between Aryl
587
Hydrocarbon Receptor and the Inflammatory Response. J. Biol. Chem. 2014, 289, (3), 1866-
588
1875.
589
(35) Lu, P.; Yan, J.; Liu, K.; Garbacz, W. G.; Wang, P.; Xu, M.; Ma, X.; Xie, W., Activation
590
of aryl hydrocarbon receptor dissociates fatty liver from insulin resistance by inducing
591
fibroblast growth factor 21. Hepatology 2015, 61, (6), 1908-19.
592
(36) Wang, C.; Xu, C.; Krager, S. L.; Bottum, K. M.; Liao, D.; Tischkau, S. A., Aryl
593
Hydrocarbon Receptor Deficiency Enhances Insulin Sensitivity and Reduces PPAR-α Pathway
594
Activity in Mice. Environ. Health Perspect. 2011, 119, (12), 1739-1744.
595
(37) Kurita, H.; Yoshioka, W.; Nishimura, N.; Kubota, N.; Kadowaki, T.; Tohyama, C., Aryl
596
hydrocarbon receptor-mediated effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on glucose-
597
stimulated insulin secretion in mice. J. Appl. Toxicol. 2009, 29, (8), 689-694.
598
(38) Roh, E.; Kwak, S. H.; Jung, H. S.; Cho, Y. M.; Pak, Y. K.; Park, K. S.; Kim, S. Y.; Lee,
599
H. K., Serum aryl hydrocarbon receptor ligand activity is associated with insulin resistance and
600
resulting type 2 diabetes. Acta Diabetol. 2015, 52, (3), 489-495.
601
(39) Trnovec, T., Dioxin relative effect potencies calculated from human thyroid data.
602
Endocrine Disruptors 2014, 2, (1), e27904.
603
(40) Nebert, D. W.; Dalton, T. P.; Okey, A. B.; Gonzalez, F. J., Role of aryl hydrocarbon
604
receptor-mediated induction of the CYP1 enzymes in environmental toxicity and cancer. J. Biol.
605
Chem. 2004, 279, (23), 23847-23850.
606
(41) van Ede, K. I.; Aylward, L. L.; Andersson, P. L.; van den Berg, M.; van Duursen, M. B.
607
M., Tissue distribution of dioxin-like compounds: Potential impacts on systemic relative
608
potency estimates. Toxicol. Lett. 2013, 220, (3), 294-302. 27 ACS Paragon Plus Environment
Environmental Science & Technology
609
(42) Carlson, E. A.; McCulloch, C.; Koganti, A.; Goodwin, S. B.; Sutter, T. R.; Silkworth, J.
610
B., Divergent Transcriptomic Responses to Aryl Hydrocarbon Receptor Agonists between Rat
611
and Human Primary Hepatocytes. Toxicol. Sci. 2009, 112, (1), 257-272.
612
(43) Silkworth, J. B.; Koganti, A.; Illouz, K.; Possolo, A.; Zhao, M.; Hamilton, S. B.,
613
Comparison of TCDD and PCB CYP1A Induction Sensitivities in Fresh Hepatocytes from
614
Human Donors, Sprague-Dawley Rats, and Rhesus Monkeys and HepG2 Cells. Toxicol. Sci.
615
2005, 87, (2), 508-519.
616
(44) Sutter, C. H.; Bodreddigari, S.; Sutter, T. R.; Carlson, E. A.; Silkworth, J. B., Analysis of
617
the CYP1A1 mRNA Dose-Response in Human Keratinocytes Indicates that Relative Potencies
618
of Dioxins, Furans, and PCBs Are Species and Congener Specific. Toxicol. Sci. 2010, 118, (2),
619
704-715.
620
(45) Westerink, W. M.; Stevenson, J. C.; Schoonen, W. G., Pharmacologic profiling of human
621
and rat cytochrome P450 1A1 and 1A2 induction and competition. Arch. Toxicol. 2008, 82,
622
(12), 909-921.
623 624
28 ACS Paragon Plus Environment
Page 28 of 36
Page 29 of 36
Environmental Science & Technology
625 626
Table 1 The calculated relative effect potencies (REPs) of PCDD/Fs and dl-PCBs with various outcomes from different human studies. REP (βi/βTCDD) calculated from different outcomes WHO CTF Congener TEF GDM FBG TV FT4 CYP1A1 CYP1B1 Human Rat
PCDDs 2378-TCDD
1
1
1
1
1
1
1
1
1
0.517
-
0.432
0.471
0.40371
-
1
1
0.5
123478-HxCDD
-
-
0.257
0.805
0.52356
-
0.1
0.03
0.2
123678-HxCDD
0.0428
0.0765
0.082
0.126
0.01386
-
0.1
0.06
0.06
123789-HxCDD
0.0526
0.0455
-
0.482
-
-
0.1
0.002
0.3
1234678-HpCDD
0.0090
0.0639
0.008
0.029
0.01943
-
0.01
0.2
0.04
-
-
0.003
-
-
-
0.0003
0.005
-
2378-TCDF
0.0805
0.0676
0.828
0.1
0.4105
2.0429
0.1
0.1
0.2
12378-PeCDF
0.441
0.288
0.347
0.03
0.6
0.2
23478-PeCDF
0.099
0.403
0.016
123478-HxCDF
-
0.0463
123678-HxCDF
0.058
0.118
234678-HxCDF
-
123789-HxCDF
12378-PeCDD
OCDD PCDFs
0.61548 0.02
0.00891
0.1657
0.3
1
0.2
-
0.05234
0.74
0.1
1
0.09
0.146
-
0.24078
0.1
0.04
0.07
-
0.78
-
0.59437
1.1857
0.1
0.06
0.07
-
-
-
-
-
-
0.1
0.02
0.3
1234678-HpCDF
0.011
0.0957
0.054
0.053
0.21088
1.1143
0.01
0.01
0.01
1234789-HpCDF
0.015
0.0412
-
-
-
-
0.01
0.3
0.05
-
-
-
0.373
0.01243
-
0.0003
0.2
0.007
PCB77
0.00076
0.0104
-
-
-
-
0.0001
-
0.0004
PCB81
0.0329
0.145
-
-
0.00093
0.2057
0.0003
-
0.0002
PCB126
0.152
0.0745
-
-
-
0.0171
0.1
0.003
0.09
PCB169
0.0362
0.102
-
-
-
0.0371
0.03
-
0.002
PCB105
-
-
-
1.8E-06
-
0.00015
0.00003
-
0.00001
PCB114
0.0114
0.0282
-
-
-
0.00067
0.00003
-
0.00006
PCB118
0.000067
0.000463
-
-
-
0.000043
0.00003
-
0.000009
PCB123
-
0.000178
-
-
-
0.00011
0.00003
-
0.000009
PCB156
0.000355
0.00195
-
-
-
0.000077
0.00003
-
0.00008
PCB157
0.00544
0.0135
-
-
-
0.000497
0.00003
-
0.00003
PCB167
0.00306
0.00579
-
-
-
0.00017
0.00003
-
0.000007
PCB189
-
0.00549
-
-
-
0.00027
0.00003
-
0.000007
OCDF DL-PCBs
627
Note: Relative effect potency (REP) was calculated as the ratio of the regression coefficients (β) of the
628
individual congener to that of TCDD. Gestational diabetes mellitus (GDM) and fasting blood glucose (FBG)
29 ACS Paragon Plus Environment
Environmental Science & Technology
629
were the outcomes investigated in present study; Thyroid volume (TV), free thyroxine (FT4), cytochrome
630
P450 (CYP) 1A1 and 1B1 expression were the outcomes investigated in European Union (EU) SYSTEQ
631
project (Trnovec et al.17 and Wimmerová et al.18). Consensus toxicity factors (CTF) were developed by
632
Larsson et al. 9.
633
30 ACS Paragon Plus Environment
Page 30 of 36
Page 31 of 36
Environmental Science & Technology
634 635
Figure 1. Correlation matrix for human studies derived REPs from GDM, FBG, CYP 1A1, CYP 1B1, thyroid
636
volume and serum FT4 outcomes (Number: Pearson correlation coefficients r; Significance level: *, P