Associations between Exposure to Persistent Organic Pollutants and

Subscriber access provided by KEAN UNIV

Ecotoxicology and Human Environmental Health

Associations between exposure to persistent organic pollutants and thyroid function in a case-control study of East China Xu Han, Lingling Meng, Yingming Li, An Li, Mary Turyk, Ruiqiang Yang, Pu Wang, Ke Xiao, Wenjuan Li, Junpeng Zhao, Qinghua Zhang, and Guibin Jiang Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b02810 • Publication Date (Web): 29 Jul 2019 Downloaded from pubs.acs.org on August 1, 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 35

Environmental Science & Technology

1

Associations between exposure to persistent organic pollutants and

2

thyroid function in a case-control study of East China

3

Xu Han1,2, Lingling Meng3, Yingming Li1*, An Li4, Mary E. Turyk4, Ruiqiang Yang1,

4

Pu Wang1, Ke Xiao1, Wenjuan Li1, Junpeng Zhao1,2, Qinghua Zhang1, 2, Guibin

5

Jiang1, 2

6 7

1State

8

Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing

9

100085, China

Key Laboratory of Environmental Chemistry and Ecotoxicology, Research

10

2University

11

3Shandong

12

First Medical University, Jinan 250014, China

13

4School

of Chinese Academy of Sciences, Beijing 100049, China

Provincial Qianfoshan Hospital, the First Hospital Affiliated with Shandong

of Public Health, University of Illinois at Chicago, Chicago, IL 60612, USA

14 15

Author information

16

Corresponding author

17

*Tel/Fax: +8610-62849818; email: [email protected]

18

Notes

19

The authors declare no competing financial interest.

1

ACS Paragon Plus Environment


20

Abstract: Animal studies have indicated that persistent organic pollutants (POPs)

21

affect thyroid hormone homeostasis, while epidemiological studies involving human

22

have not shown consistent results. In this study, we investigated the associations

23

between POP exposure and thyroid function among adult population of East China.

24

One hundred and eighty-six participants diagnosed with thyroid disease and 186

25

participants without thyroid disease from Shandong, China were enrolled in the case-

26

control study during 2016-2017. We found that POP exposure was significantly and

27

positively associated with the risk of thyroid disease. The association of thyroid disease

28

with the sum of 17 POPs followed a non-monotonic dose response, with an adjusted

29

odds ratio (OR) of 2.09 [95% confidence intervals (CI): 1.13-3.87, p=0.019] for the

30

second quartile. Among 186 participants in the control group, concentrations of POPs

31

showed negative associations with triiodothyronine (T3), free T3 (FT3), thyroxine (T4)

32

and free T4 (FT4) in males and positive associations with T4 and FT4 in females. Taken

33

together, these findings suggest that POP exposure can disrupt thyroid hormone

34

homeostasis and increase the risk of thyroid disease.

2


Page 2 of 35

Page 3 of 35


35

Introduction

36

Thyroid hormones play critical roles in metabolism and growth, especially in the

37

development of the neonatal brain1. Thyroid disease is a common health problem

38

worldwide2. As a common endocrine malignancy, thyroid cancer has been the fastest

39

growing cancer worldwide over the past decades3, 4. Thyroid cancer is more common

40

in females, and accounts for 5.1% of the total estimated cancer burden for females. In

41

2018, the global incidence rate of thyroid cancer was 10.2 per 100,000 in females, and

42

3.1 per 100,000 in males5. Hypothyroidism, the most common thyroid disorder, is

43

associated with cardiovascular disease and metabolic syndrome and other adverse

44

health effects6-8. Hyperthyroidism, another common thyroid dysfunction, is also

45

associated with many complications, such as increased cardiovascular risk, atrial

46

fibrillation and thyrotoxic periodic paralysis9, 10. The rapid increase in the incidence and

47

potentially adverse health outcomes of thyroid disease has caused increasing public

48

concern in recent years.

49

Although factors such as better diagnostics, inadequate iodine nutrition, and

50

radiation exposure may play important roles in the increasing incidence of thyroid

51

disease, exposure to environmental chemicals also contributes through their effects on

52

thyroid gland morphology and thyroid hormone homeostasis2, 11, 12. Persistent organic

53

pollutants (POPs), a group of lipophilic chemicals resistant to degradation, can

54

accumulate in the human body and cause adverse health effects13, 14. A wide range of

55

experimental studies in animals and in vitro have suggested that exposure to POPs is 3



56

associated with disruption of thyroid hormone homeostasis15-18. POPs may interfere

57

with the natural production, transport, and metabolism of thyroid hormones through

58

various mechanisms, including inhibition of thyroid-stimulating hormone (TSH)

59

receptor, inhibition of deiodinase, displacement of thyroxine (T4) from transthyretin

60

(TTR), upregulation of liver uridine diphosphate glucuronosyltransferases (UDPGTs),

61

inhibition of sulfotransferases (SULTs), and direct action on thyroid hormone receptor

62

(TR)19-22. For example, polybrominated diphenyl ethers (PBDEs) and hydroxylated-

63

PBDEs (OH-PBDEs) have highly similar chemical structures to T4 and may compete

64

with binding to TTR, possibly causing a decrease in free T4 (FT4) concentration in

65

serum. In particular, competitive binding of PBDEs and hydroxylated-PBDEs may lead

66

to a possible effect on the fetal brain because TTR may mediate the delivery of T4

67

across the blood-brain barrier and facilitate maternal to fetal transport through the

68

placenta21-23. Dioxin-like polychlorinated biphenyls (DL-PCBs) can bind to the aryl

69

hydrocarbon receptor (AhR) and therefore may induce UDPGTs, resulting in increased

70

glucuronidation and excretion of T4. Non-DL-PCBs that do not bind to AhR may

71

disrupt thyroid hormone homeostasis through AhR-independent mechanisms, such as

72

inducing cytochrome P450 (CYP) 2B and 3A enzymes or displacing T4 from TTR-like

73

PBDEs and OH-PBDEs24-27. Certain organochlorine pesticides (OCPs), such as

74

dichlorodiphenyltrichloroethane (DDT), hexachlorobenzene (HCB) and chlordane,

75

have also been suggested to disrupt thyroid hormone homeostasis in animals and

76

humans22. 4


Page 4 of 35

Page 5 of 35


77

Epidemiological studies have also focused on the associations between POP

78

exposure and thyroid hormones in humans. A previous study including 72 occupational

79

workers from a deca-BDE manufacturing plant found that a 10-fold increase in serum

80

concentrations of BDE-209 was associated with an 8.63 nmol/L increase in total T4

81

(TT4) [95% confidence intervals (CI): 0.930 to 16.3] and 0.106 nmol/L increase in total

82

triiodothyronine (TT3) (95% CI:-0.005 to 0.219)28. Similar associations were also

83

observed between maternal PBDE exposure and thyroid hormones. Vuong et al. found

84

that concentrations of BDE-28 and 47 in the serum of pregnant women were positively

85

associated with maternal TT4 and FT4 but were not associated with thyroid hormone

86

levels in cord serum29. However, animal studies have always shown inverse

87

relationships between POP exposure and triiodothyronine (T3) or T4. Significant

88

negative associations between PCBs 138, 153, 180, and 187 and free T3 (FT3) were

89

found in a study that included 320 children 7-10 years of age. PCB-118 was positively

90

associated with TSH, and no PCB congener showed a significant association with FT430.

91

Although the associations between POP exposure and thyroid hormones might not be

92

comparable among occupationally exposed workers, pregnant women, and children,

93

the results have been inconsistent even in the same population group. Therefore, more

94

epidemiological studies are needed to investigate the effect of POPs on the disruption

95

of thyroid hormones.

96

At present, relatively few studies have focused on the relationships between POPs

97

exposure and thyroid disease31-33. To our knowledge, no previous study has reported 5



98

the associations between POP exposure and thyroid disease in China, a country with a

99

massive population and an increasing incidence of thyroid cancer34, 35. In 2014, the age-

100

standardized incidence rate by world standard population of thyroid cancer was

101

9.29/100000 in China, and 4.67/100000 for male, 14.05/100000 for female36. In

102

particular, East China had the largest estimated number of new cases of thyroid cancer

103

(40.2 thousand cases) in China in 2015, followed by Central China (14.3 thousand cases)

104

and South China (11.2 thousand cases)37. Therefore, the main aims of the present study

105

were to (1) evaluate the association between POP exposure and thyroid disease risk in

106

Chinese adults, and (2) analyze the associations between POP concentrations in human

107

serum and thyroid hormones in healthy people.

108 109

Materials and Methods

110

Sample collection. Participants were enrolled from Jinan, the capital city of

111

Shandong Province in China, during 2016-2017. A total of 447 people (212 in the case

112

group and 235 in the control group) were initially enrolled. Of the 212 participants in

113

the case group, 13 who had missing data on triglycerides or total cholesterol and 13

114

who had missing samples were excluded. All patients with thyroid disease had been

115

identified in Shandong Provincial Qianfoshan Hospital in Jinan, and patients with

116

thyroid cancer underwent thyroidectomy. Patients with thyroid cancer was diagnosed

117

by surgical pathology or fine needle aspiration cytology. Hyperthyroidism is

118

characterized by increased serum concentrations of thyroid hormones (T3, T4, FT3 and 6


Page 6 of 35

Page 7 of 35


119

FT4), and decreased serum concentrations of TSH. On the contrary, hypothyroidism is

120

characterized by decreased serum concentrations of thyroid hormones (T3, T4, FT3 and

121

FT4), and increased serum concentrations of TSH. Other types of thyroid disease such

122

as subacute thyroiditis were diagnosed through different ways according to the stage of

123

the disease.

124

A total of 235 participants in the control group were selected from people who

125

underwent a physical examination in the same hospital. People were excluded from the

126

control group if they had a history of physician-diagnosed thyroid disease. Efforts were

127

made to randomly match the control group with the case group for age (+/-5 years) and

128

residential area. Finally, 186 participants enrolled in the control group. All participants

129

donated their blood samples and provided information about age and sex. Participants

130

with diabetes were identified if they had a history of physician-diagnosed diabetes or

131

had a fasting plasma glucose level more than 7.0 mmol/L. Patients with thyroid disease

132

did not use thyroid medications for at least 2 weeks before blood collection, as directed

133

by their doctors for future diagnosis and treatment. Fasting blood was collected from

134

all participants.

135

POPs analysis. 13C-labeled standard mixture of PCBs (68C-LCS and 68C-IS) and

136

PBDEs (MBDE-MXG-LCS) were purchased from Wellington Laboratories (Ontario,

137

Canada). 13C-labeled standard mixture of OCPs (ES-5465 and EC-5350) were obtained

138

from Cambridge Isotope Laboratories (Andover, MA, USA). Solvents dichloromethane

139

(DCM), n-hexane and methanol are pesticide grade, and were from J.T. Baker 7



140

Company Inc. (Fairfield, OH, USA). 2-Propanol and formic acid (99%) were bought

141

from Merck (Darmstadt, Germany) and Acros Organics (Belgium), respectively. Silica

142

gel 60 (0.063-0.100 mm) and anhydrous sodium sulfate (Na2SO4), obtained from Merck,

143

were activated at 550C for 12 h and 660C for 6 h prior to use, respectively.

144

The collected serum samples were stored at -20C until chemical analysis for

145

POPs. Methods for sample preparation were adopted from previously published

146

procedures38, 39. In brief, 13C-labeled surrogate standard mixtures were spiked into 0.5

147

mL of each serum sample prior to extraction. Solid phase extraction (SPE) using an

148

Oasis® HLB cartridge (6 cm3/500 mg, Waters, Milford, MA, USA) was used to extract

149

the POPs from the serum samples. A small multilayer silica gel column containing

150

layers of 33% H2SO4 silica gel, neutral silica gel, and active anhydrous Na2SO4 was

151

used for the cleanup of the extract. The targeted POPs were eluted by 10 mL of n-

152

hexane. The concentrated elute was spiked with 13C-labeled injected standards (68C-IS

153

and EC-5350) for instrumental analysis. A total of 22 PCBs (PCB-77, 81, 105, 114,

154

118, 123, 126, 156, 157, 167, 169, 189, 28, 52, 101, 138, 153, 180, 209, 202, 205 and

155

208) and a total of 27 PBDEs (BDE-3, 7, 15, 17, 28, 47, 49, 66, 71, 77, 85, 99, 100,

156

119, 126, 138, 153, 154, 156, 183, 184, 191, 196, 197, 206, 207, 209) in serum were

157

analyzed by high-resolution gas chromatography coupled with high-resolution mass

158

spectrometry (HRGC/HRMS, AutoSpec Premier, Waters, USA). A total of 26 OCPs

159

(pentachlorbenzene, α-HCH, HCB, β-HCH, γ-HCH, δ-HCH, heptachlor, aldrin, oxy-

160

chlordane, cis-heptachlor epoxide, trans-heptachlor epoxide, trans-chlordane, o,p’8


Page 8 of 35

Page 9 of 35


161

DDE, cis-chlordane, endosulfan I, trans-nonachlor, dieldrin, p,p’-DDE, o,p’-DDD,

162

endrin, endosulfan II, cis-nonachlor, p,p’-DDD, o,p’-DDT, p,p’-DDT and

163

mirex/kepone) were determined using another HRGC/HRMS (DFS, Thermo Fisher,

164

USA). Details about the instrumental analysis of POPs are described elsewhere40, 41.

165

A procedural blank was added to each analytical batch of 11 samples. Most

166

compounds were not detected in blank samples. PCB-118, 138, 153 and p,p’-DDE were

167

detected in the blank samples with very low concentrations (less than 10% of the

168

concentration in samples). Reported concentrations were not blank corrected. The

169

recoveries of the spiked surrogate standards in serum samples were 50-129%, 41-114%

170

and 37-126% for PCBs, PBDEs, and OCPs, respectively. The limit of detection (LOD)

171

was defined as three times signal/noise ratio. In this study, the LODs were 0.03-1.39

172

pg/mL for PCBs, 0.05-3.16 pg/mL for PBDEs and 0.05-9.75 pg/mL for OCPs.

173

Clinical chemistry parameter analysis. Clinical chemistry parameter analysis

174

was performed in the hospital. The reference range for thyroid hormones in healthy

175

adults was 3.10-6.80 pmol/L for FT3, 12.00-22.00 pmol/L for FT4, 1.30-3.10 nmol/L

176

for T3, 66.00-181.00 nmol/L for T4 and 0.27-4.20 µIU/mL for TSH. Total lipids (TLs)

177

were calculated by the short formula TL=2.27×total cholesterol+triglycerides+0.62342.

178

Statistical analysis. Mann-Whitney U tests were used to assess the differences in

179

POP concentrations, triglyceride levels and total lipids between the case and control

180

groups. To compare the differences in age and total cholesterol between cases and

9



181

controls, Student’s t-tests were employed. Chi-square tests were used to examine the

182

differences in sex and diabetes status between the two groups.

183

We used Spearman’s rank correlations to evaluate the associations of POP

184

concentrations with age, triglycerides, total cholesterol and thyroid hormones. Both

185

wet-weight based and lipid-standardized concentrations of POPs were evaluated.

186

Logistic regression models were used to estimate the associations between POPs in

187

serum and the risk of thyroid disease by calculating crude and adjusted odds ratios (ORs)

188

and their corresponding 95% CIs. Based on the distributions among the controls,

189

concentrations of POPs were categorized into quartiles. The reference category was the

190

lowest exposure category. P-values for linear trend were estimated based on the median

191

value of each category of POPs. A sequential method of controlling for confounding

192

was used, with Model 1 controlling for age and sex, model 2 controlled for diabetes

193

status in addition to Model 1 covariates, and Model 3 additionally controlled for

194

triglycerides and total cholesterol. We did not control for body mass index (BMI)

195

because body weight may be affected by thyroid function43-45. Even slightly increased

196

serum TSH levels are associated with high occurrence of weight gain45, 46.

197

The occurrence of thyroid diseases was examined for the potential links to the

198

serum concentrations of total PCBs (∑PCBs), total dioxin-like PCBs (∑DL-PCBs),

199

total non-dioxin-like PCBs (∑NDL-PCBs), total PBDEs (∑PBDEs), total OCPs

200

(∑OCPs), and total POPs (∑POPs). ∑DL-PCBs is the concentration sum of PCB-105,

201

114, 118, 156, 157, and 167; and ∑NDL-PCBs is the concentration sum of PCB-138, 10


Page 10 of 35

Page 11 of 35


202

153, and 180. Considering that females had a higher prevalence of thyroid disease, we

203

stratified the population in the current study by sex. Additionally, the associations

204

between POPs and thyroid cancer were evaluated. Among participants in the control

205

group, multiple linear regressions were run to assess relationships between POP

206

concentrations and thyroid hormones. Age, sex and diabetes status were included in the

207

final multivariate models as covariates. A sex-stratified analysis was also presented in

208

the regression to evaluate the joint effect of POPs and sex.

209

Values below the LOD were replaced by 1/2 LOD. Statistical significance was set

210

at p75%

232

of the serum sample; PCB-114 was detected in 73% of serum samples. PCB-138, 153

233

and 180 were detected in all serum samples. PCB-153 was the dominant PCB congener,

234

and the median concentrations were 5.33 and 5.25 ng/g in the case and control groups,

235

respectively (Table S2). The geometric mean (GM ± standard deviation) concentration

236

of BDE-153 was 1.10±2.15 ng/g, higher than that of BDE-47 (0.13±2.20 ng/g). GM

237

concentration of p,p’-DDE was 290.47±3.01 ng/g, which is orders of magnitude higher

238

than the GMs for most measured POPs. p,p’-DDE is followed by β-HCH, which also

239

stands out having the second highest GM of 77.17±3.10 ng/g. Tables S3-S5 show the

240

comparison of serum POP concentrations in this study with recent researches from

241

different regions. Compared with other population, serum PCB concentrations in this

242

study were generally lower, while concentrations of PBDEs and OCPs were in

12


Page 12 of 35

Page 13 of 35


243

moderate levels. The concentrations of all the 17 POPs were natural log-transformed

244

for normalization.

245

The concentrations of POP congeners such as PCB-114, BDE-47, 153, trans-

246

chlordane and p,p’-DDT were significantly higher in the case group than in the control

247

group (Table S2), while PCB-156 and mirex/kepone were slightly lower in the case

248

group than in the control group. Other POP congeners did not show significant

249

difference between the case and the control group. The concentrations of PCB-156,

250

157, 167, 138, 153 and 180, BDE-153 and mirex/kepone were significantly higher in

251

males than in females (data not shown). Other POPs showed no significant difference

252

by sex. In addition to BDE-47, trans-chlordane and mirex/kepone, other POP congeners

253

were significantly positively associated with age (data not shown).

254

Associations between POPs and thyroid disease. Logistic regression analyses

255

showed that lipid-standardized concentrations of most POPs were positively associated

256

with thyroid disease (Table 1). After adjusting for age, sex and diabetes status in the

257

model, PCB-105, 114 and 157, BDE-47, 153, trans-chlordane and p,p’-DDE showed

258

significantly positive associations with thyroid disease. However, the associations of

259

PCB-156, 138, 180 and mirex/kepone with thyroid disease were significantly negative.

260

For example, the ORs in the final model ranged from 0.23 to 0.71 for PCB-156, and

261

from 0.27 to 0.56 for mirex/kepone (Table 1). In addition to ∑OCPs and ∑POPs, the

262

associations between mixtures of POPs and thyroid disease failed to reach significance

263

after adjusting for age, sex and diabetes status. ∑OCPs and ∑POPs showed 13



264

significantly positive associations with thyroid disease, with the adjusted ORs for the

265

second quartile of 1.97 (95% CI: 1.07-3.63) and 2.09 (95% CI: 1.13-3.87), respectively

266

(Table 1). The dose-response relationships between POPs and thyroid disease were not

267

consistent for all POPs (Table 1). BDE-47, 153 and trans-chlordane showed monotonic

268

and positive associations with thyroid disease, while some PCB congeners such as

269

PCB-156 and 180 showed monotonic and negative associations. For example, the ORs

270

for BDE-47 were 1.24, 1.12 and 2.60 for the second, third, and fourth quartiles,

271

respectively, and the linear regression had a p-value of 0.001. The ORs were 0.71, 0.51

272

and 0.23 for the second, third, and fourth quartiles for PCB-156, with the p-value less

273

than 0.001. In addition, some POP congeners presented nonlinear associations with

274

thyroid disease. For example, p,p’-DDE showed an inverse U shape in its associations

275

with thyroid disease. The OR for the second quartile was 2.07, showing the strongest

276

association with the risk of thyroid disease, while the ORs were 1.38 for the third

277

quartile and 1.23 for the fourth quartile (p-value = 0.564 for the linear trend).

278

To account for the higher prevalence of thyroid disease in females, sex-stratified

279

models were used (Table S6). The results in females were consistent with the overall

280

results. PCB-114, 157, BDE-47, trans-chlordane and p,p’-DDT were significantly

281

positively associated with thyroid disease in females, whereas PCB-118, 156, 180 and

282

mirex/kepone showed negative associations. However, only the associations between

283

PCB-153 and mirex/kepone and thyroid disease reach statistical significance in males.

14


Page 14 of 35

Page 15 of 35


284

When we restricted our analyses to participants with thyroid cancer, the results

285

were similar to the main analysis that included both non-cancerous thyroid disease and

286

thyroid cancers (Table S7). For example, PCB-114 in the serum was positively

287

associated with thyroid disease, with the adjusted ORs for the third quartile of 2.82 (95%

288

CI: 1.52-5.23) for the all kinds of thyroid disease, and 2.38 (95% CI: 1.24-4.57) for

289

thyroid cancer.

290

Concerning the effect of lipids on the associations between POPs and thyroid

291

disease, both wet-weight concentrations of POPs, adjusted for triglycerides and total

292

cholesterol, and lipid-standardized concentrations were used in the analysis (Table S8).

293

However, the results were similar for both, so we presented only the lipid-standardized

294

results in this paper, to facilitate the comparisons with other studies.

295

Associations between POPs and thyroid hormones. Among 186 participants in

296

the control group, T3 and FT3 were significantly and negatively correlated with age

297

(p

Associations between Exposure to Persistent Organic Pollutants and

Recommend Documents