Iodide-Induced Chemokines and Genes Related to Immunological Function in Cultured Human Thyroid Follicles in the Presence of Thyrotropin

Abstract

Background:

It is well known that iodide exacerbates thyroid function in subclinical hypothyroid patients with autoimmune thyroiditis. To investigate the immunological mechanism of iodine-induced thyroid dysfunction, we studied the effect of iodide in cultured human thyroid follicles, which respond to physiological concentrations of human thyrotropin (TSH) (0.3–10 μU/mL) and maintain the Wolff–Chaikoff effect.

Materials and Methods:

Thyroid follicles obtained from Graves' patients at subtotal thyroidectomy were precultured in medium containing 0.5% fetal calf serum and 10⁻⁸ M iodide for 5 days, and then cultured with the medium containing bovine TSH (30 μU/mL) and low (10⁻⁸M) or high (10⁻⁵M) concentrations of iodide. After 3–72 hours of culture, the effect of iodide on thyroid cell mRNA expression was analyzed by microarray and reverse transcriptase-polymerase chain reaction.

Results:

After 48 hours of culture, iodide nearly doubled the mRNA expression levels of the immunity-associated genes (intercellular adhesion molecule-1, transforming growth factor beta 1–induced protein, early growth response gene 1, guanylate-binding protein 1, and annexin A1) and decreased the mRNA expression of sodium-iodide symporter to less than 20%. Further, the mRNA expression levels of chemokines (CCL2, CXCL8, and CXCL14) increased nearly twofold, whereas their receptors did not show any significant response. Real-time polymerase chain reaction analyses confirmed that iodide increased the mRNA expression levels of these genes in a time- and concentration-dependent manner. Immunohistochemical studies revealed that the chemokines were expressed mainly in the thyroid follicular cells in addition to the immune cells. The iodide-induced increase in CCL2 was greater in thyroid follicles obtained from thyroid gland that had been moderately infiltrated with the immunocompetent cells.

Conclusion:

We have demonstrated that iodide stimulates thyroid follicular cells to produce chemokines, particularly CCL2, CXCL8, and CXCL14. These chemokines and intercellular adhesion molecule-1 would attract immunocompetent cells into thyroid gland. These in vitro findings suggest that iodide at high concentrations may induce thyroid dysfunction through not only biochemical but also immunological mechanisms, particularly in patients with autoimmune thyroid disorders.

Introduction

I t is well known that patients with autoimmune thyroid disease (particularly, Hashimoto's thyroiditis and radiologically or surgically treated Graves' disease) are prone to develop hypothyroidism after exposure to a large dose of iodine (1 –4). Further, when the thyroid gland in Graves' patients is small and the thyrotropin (TSH) receptor antibody titer is low, a large dose of iodide may induce euthyroidism for a prolonged period (5). To investigate the immunological mechanism of iodine-induced thyroid dysfunction, we studied the effect of iodide on chemokines and cytokines in cultured human thyroid follicles that contain a minimal number of lymphocytes. Thyroid follicles in suspension culture incorporate ¹²⁵I and release de novo synthesized ¹²⁵I-T3 and ¹²⁵I-T4 into the culture medium at a physiological concentration of human TSH (0.3–10.0 μU/mL) (6) and maintain the Wolff–Chaikoff effect (7).

Chemokines are a family of small, structurally related molecules that regulate the trafficking of various types of immune cells through interactions with their receptors (8,9). Their major function is the recruitment of immunocompetent cells to sites of inflammation, but they also play roles in angiogenesis and autoimmunity. To date, 47 chemokines and 20 chemokine receptors have been identified in humans (10).

Previously, we demonstrated that iodide increased the levels of expression of mRNA for intercellular adhesion molecule-1 (ICAM1) and CXCL8 (interleukin [IL]-8, monocyte-derived neutrophil chemotactic factor [MDNCF]) in cultured human thyroid follicles using a microarray for 2400 genes (11). In this study, using an oligo-DNA microarray capable of analyzing all human genes (41,000 spots) in a single run (12), we investigated the effect of iodine on genes related to immunological actions, particularly chemokines and their receptors, although it is well known that several chemokines (CXCL8, CCL3, CCL4, CXCL1, CXCL12, CCL21) are to some extent overexpressed in thyroid gland in Graves' patients (13 –16).

Materials and Methods

Suspension culture of human thyroid follicles

This study was approved by the Ethics Committee of thr Tokyo Women's Medical University. Informed consent was obtained from all patients with Graves' disease before subtotal thyroidectomy. All the patients were routinely treated with iodide (20–40 mg/day) and antithyroid agents (usually methimazole) for 7–10 days before surgery. Human thyroid follicles were prepared and cultured as reported previously (6). The washed thyroid follicles were resuspended in 15 mL of F-12/RPMI-1640 (1:1) medium supplemented with 0.5% fetal bovine serum and 10⁻⁸ M NaI (standard medium) (about 1000–2000 follicles/mL) and were added to 10-cm dishes, the bottoms of which had been coated with agarose. Although the thyroid gland in Graves's disease is minimally to moderately infiltrated by immunocompetent cells, in our thyroid follicle suspension culture system, most of such cells were removed from the follicles by repeated centrifugation, although a few immune cells that had infiltrated intraepithelially could not be removed (17).

RNA extraction

Total RNA was extracted from thyroid follicles using the RNeasy Minikit (Quiagen, Tokyo, Japan) in accordance with the manufacturer's suggested protocol, as described previously (18).

Oligo-DNA microarray

Oligo-DNA microarray (Whole Human Genome OligoMicroarray Kit [Agilent Technologies, Palo Alto, CA]) was performed as described previously (18).

Laser detection of the Cy3 (green, NaI 10⁻⁸M) and Cy5 (red, NaI 10⁻⁵M) signals on the microarray was performed with a dual-laser DNA microarray scanner (Agilent Technologies). Fluorescence signal intensities and the Cy5/Cy3 ratios for each of the 41,000 oligo-DNAs were analyzed using the Feature Extraction software package (Agilent Technologies). The microarray data discussed in this publication have been deposited in the Gene Expression Omnibus (GEO; www.ncbi.nlm.nih.gov/geo/info/linking.html) of the National Center for Biotechnology Information and are accessible through the GEO series accession no. GSE16957). The microarray data for the effect of TSH on thyroid follicles have been deposited in the GEO (accession no. GSE12244) (12).

Real-time reverse transcriptase–polymerase chain reaction

Real-time reverse transcriptase–polymerase chain reaction (RT-PCR) was performed as described previously (18). Nine assays were performed using 18S rRNA (Hs99999901_s1) as a control: ICAM-1 (Hs00164932_m1), transforming growth factor beta-1 (TGF-β1) (Hs00171257_m1), TGF-β-induced protein (TGFBI, Hs00165908_m1), CCL2 (Hs00234140_m1), IL-6 (Hs00174131_m1), SLC5A5 (sodium-iodide symporter [NIS], Hs00166567_m1), CXCL8 (IL-8, Hs00174103_m1), CXCL14 (Hs00171135_m1), and Annexin A1 (ANXA1, Hs00945401_m1). The PCR thermal cycle conditions were set at 50°C for 2 minutes and 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute.

Immunohistochemistry

Graves' thyroid tissues were fixed in 10% neutral-buffered formalin for 24 hours, then embedded in paraffin as a routine procedure. A modified antigen-retrieval method (18) was used for the immunohistochemical observation of CCL2, CXCL8, CXCL14, and CXCR7. Briefly, sections were placed in a plastic Coplin jar filled with 0.001N NaOH and heated for 5 minutes at 120°C in an autoclave. The sections were then exposed to 3% H₂O₂ to inactivate endogenous peroxidase, followed by incubation overnight at 4°C with anti-CCL2 (R&D Systems, Minneapolis, MN), anti-CXCL8 (Lifespan Biosciences, Seattle, WA), anti-CXCL14 (Proteintech Group, Chicago, IL), or anti-CXCR7 (Lifespan Biosciences) antibodies at 1:200 dilution. The labeled streptavidin–biotin method (LSAB + System HRP kit; Dako, Glostrup, Denmark) was used in accordance with the manufacturer's protocol. 3,3′-Diaminobenzidine tetrahydrochloride was used to detect peroxidase activity. Sections were lightly counterstained with hematoxylin for 10 seconds to view nuclei.

Statistical analysis

All data are expressed as mean ± SD. Differences were considered significant at p < 0.05.

Results

Genes upregulated by iodide in cultured human thyroid follicles

When thyroid follicles were cultured in high iodide medium for 3–6 hours, microarray analyses revealed little changes in the mRNA expression levels compared with the follicles cultured for 48–72 hours (Fig. 1A). Therefore, microarray analyses were performed for 48 hours in five independent experiments.

FIG. 1.

Effect of iodide on gene expression levels of chemokines (scatter plot) (A) and their receptors. (B) Thyroid follicles were precultured for 5 days and then cultured in medium containing bovine TSH (30 μU/mL) and a low (10⁻⁸ M) or high (10⁻⁵ M) iodide concentration. After an additional 2 days of culture, total RNA was extracted, and the mRNA expression levels were analyzed by oligo-DNA microarray, as described in the Materials and Methods section. TG, thyroglobulin; TPO, thyroperoxidase; ANXA1, annexin A1; TGFβ1, transforming growth factor-β1; TGFBI, TGF-beta-induced protein; ICAM1, intercellular adhesion molecule-1; NIS, sodium-iodide symporter; EGR1, early growth response gene 1; GBP1, guanylate binding protein 1; TSH, thyrotropin. Color images available online at www.liebertonline.com/thy.

The genes expressed most abundantly in cultured human thyroid follicles were thyroglobulin and thyroid peroxidase (12). When thyroid follicles were cultured in medium containing a high concentration of iodide (10⁻⁵M), about 30 genes were constantly upregulated more than twofold (Table 1A). As reported previously (12), the expression of urokinase-type plasminogen activator doubled in the iodide-treated follicles. Iodide increased the mRNA expression level of ICAM-1 more than twofold in four of the five experiments (Table 2) (12,19 –21). It also increased the mRNA expression level of TGF-β1 more than 40% in three of the five experiments (11,22,23) and that of TGF-β2 by nearly 50%, leading to a significant increase in the mRNA expression level of TGFBI (Table 1A). Further, genes at least partly involved in immunological function, such as ANXA1 (24), early growth response gene 1 (25), and guanylate-binding protein 1 (26), were also increased more than twofold (Table 1A).

Table 1A.

Genes Upregulated by Iodide More Than Twofold

Gene name	Systematic name	Description	Mean	SD
PMP22	NM_000304	Peripheral myelin protein 22, transcript variant 1	4.09	1.61^a
ZBED2	NM_024508	Zinc finger, BED domain containing 2	3.19	0.57^a
IGFBP5	NM_000599	Insulin-like growth factor-binding protein 5	2.99	0.64^a
KLF2	NM_016270	Homo sapiens Kruppel-like factor 2 (lung) (KLF2)	2.94	1.09^a
CAV1	NM_001753	Caveolin 1, caveolae protein, 22 kDa (CAV1)	2.91	0.47^a
DCAMKL1	NM_004734	Doublecortin and CaM kinase-like 1 (DCAMKL1)	2.86	0.72^a
CTGF	NM_001901	Connective tissue growth factor (CTGF)	2.84	0.75^a
NR3C2	NM_000901	Nuclear receptor subfamily 3, group C, member 2	2.81	0.41^a
TACSTD2	NM_002353	Tumor-associated calcium signal transducer 2	2.81	0.49^a
STMN1	NM_005563	Stathmin 1/oncoprotein 18 (STMN1), transcript variant 3	2.71	0.40^a
COL8A2	NM_005202	Collagen, type VIII, alpha 2 (COL8A2)	2.66	0.60^a
AHNAK	NM_001620	AHNAK nucleoprotein (desmoyokin), transcript variant 1	2.65	0.60^a
PHLDB2	NM_145753	Pleckstrin homology-like domain, family B, member 2	2.56	0.66^a
NCOA7	NM_181782	Nuclear receptor coactivator 7	2.46	0.60^a
CAV3	NM_001234	Caveolin 3, transcript variant 2	2.42	0.53^a
COPEB	NM_001300	Core promoter element-binding protein	2.42	0.49^a
PRSS23	NM_007173	Protease, serine, 23	2.40	0.25^a
ANXA1	NM_000700	Annexin A1	2.38	0.51^a
EGR1	NM_001964	Early growth response 1	2.37	0.86^a
YWHAH	NM_003405	Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, eta polypeptide	2.35	0.27^a
PLAU	NM_002658	Plasminogen activator, urokinase	2.31	0.35^a
ARG2	NM_001172	Arginase, type II, nuclear gene-encoding mitochondrial protein	2.29	0.31^a
CYR61	NM_001554	Cysteine-rich, angiogenic inducer, 61	2.24	0.59^a
LGALS3	NM_002306	Lectin, galactoside-binding, soluble, 3 (galectin 3)	2.20	0.23^a
ARNTL	NM_001178	Aryl hydrocarbon receptor nuclear translocator-like	2.16	0.24^a
DPYSL3	NM_001387	Dihydropyrimidinase-like 3	2.16	0.50^a
VIM	NM_003380	Vimentin	2.14	0.35^a
LRP4	AB011540	mRNA for MEGF7, partial cds.	2.10	0.39^a
TM4SF1	NM_014220	Transmembrane 4 superfamily member 1	2.10	0.37^a
COPEB	NM_001300	Core promoter element-binding protein	2.07	0.39^a
TGFBI	NM_000358	Transforming growth factor, beta-induced, 68 kDa	2.04	0.76^a
GBP1	NM_002053	Guanylate-binding protein 1, interferon-inducible, 67 kDa	2.02	0.32^a

Data are mean ± SD (n = 5).

p < 0.05 10⁻⁸ M versus 10⁻⁵ M NaI.

SD, standard deviation.

Table 1B.

Genes Downregulated by Iodide to Less Than 50%

Gene name	Systematic name	Description	Mean	SD
ATP5G1	NM_005175	ATP synthase, H + transporting, mitochondrial F0 complex, subunit c (subunit 9), isoform 1	0.50	0.11^a
EFHD2	NM_024329	EF hand domain containing 2	0.48	0.09^a
LDLR	NM_000527	Low-density lipoprotein receptor (familial hypercholesterolemia)	0.48	0.09^a
EBP	NM_006579	Emopamil-binding protein (sterol isomerase)	0.47	0.15^a
MTCH1	NM_014341	Mitochondrial carrier homolog 1 (C. elegans), nuclear gene-encoding mitochondrial protein	0.47	0.08^a
CRABP1	NM_004378	Cellular retinoic-acid-binding protein 1	0.44	0.03^a
POR	NM_000941	P450 (cytochrome) oxidoreductase	0.43	0.07^a
SLC16A3	NM_004207	Solute carrier family 16 (monocarboxylic acid transporters), member 3	0.43	0.06^a
CKB	NM_001823	Creatine kinase, brain	0.43	0.07^a
IDI1	NM_004508	Isopentenyl-diphosphate delta isomerase	0.43	0.12^a
CCDC3	NM_031455	Coiled-coil domain containing 3	0.39	0.17^a
EFHD2	NM_024329	EF hand domain containing 2	0.39	0.04^a
SLC5A5	NM_000453	Carrier family 5 (sodium iodide symporter), member 5	0.38	0.16^a
FADS2	NM_004265	Fatty acid desaturase 2	0.36	0.16^a
C6orf176	BC008632	Chromosome 6 open reading frame 176, mRNA (cDNA clone IMAGE:3464195), partial cds	0.35	0.08^a
AQP5	BC032946	Aquaporin 5, mRNA (cDNA clone MGC:33163 IMAGE:5269384), complete cds.	0.27	0.09^a
SCD	NM_005063	Stearoyl-CoA desaturase (delta-9-desaturase)	0.24	0.12^a

Data are mean ± SD (n = 5).

p < 0.05 10⁻⁸ M versus 10⁻⁵ M NaI.

Table 2.

Effect of Iodine on mRNA Expression Levels of Cytokines in Cultured Human Thyroid Follicles

Gene name	Systematic name	Description	Mean	SD
IL-1A	NM_000575	Interleukin 1, alpha	1.19	0.67
IL-1B	NM_000576	Interleukin 1, beta	1.44	1.28
IL-2	NM_000586	Interleukin 2	1.01	0.60
IL-6	NM_000600	Interleukin 6 (interferon, beta 2)	1.48	0.43
IL-16	NM_004513	Interleukin 16 (lymphocyte chemoattractant factor)	1.07	0.58
TNF	NM_000594	Tumor necrosis factor (TNF superfamily, member 2)	1.18	0.31
TNFAIP1	NM_021137	Tumor necrosis factor, alpha-induced protein 1 (endothelial)	1.19	0.17
TNFAIP2	NM_006291	Tumor necrosis factor, alpha-induced protein 2	0.93	0.06
TNFAIP3	NM_006290	Tumor necrosis factor, alpha-induced protein 3	1.20	0.39
TNFAIP6	NM_007115	Tumor necrosis factor, alpha-induced protein 6	1.85	1.25
TNFAIP8	NM_014350	Tumor necrosis factor, alpha-induced protein 8	1.09	0.15
IFNA2	NM_000605	Interferon, alpha 2	1.00	0.53
IFNA5	NM_002169	Interferon, alpha 5	0.89	0.49
IFNA6	NM_021002	Interferon, alpha 6	1.05	0.52
IFNA8	NM_002170	Interferon, alpha 8	0.79	0.47
IFNA10	NM_002171	Interferon, alpha 10	0.98	0.43
IFNA14	NM_002172	Interferon, alpha 14 (IFNA14)	0.96	0.63
IFNA21	NM_002175	Interferon, alpha 21	1.00	0.47
IFNB1	NM_002176	Interferon, beta 1, fibroblast	0.78	0.35
IFNG	NM_000619	Interferon, gamma	0.76	0.46
TGFβ1	NM_000660	Transforming growth factor, beta 1 (Camurati–Engelmann disease)	1.22	0.08
TGFβ2	NM_003238	Transforming growth factor, beta 2	1.46	0.44
TGFβ3	NM_003239	Transforming growth factor, beta 3	0.83	0.23^a
Others
ICAM1	NM_000201	Intercellular adhesion molecule 1 (CD54), human rhinovirus receptor	1.91	0.41^a
TGFBI	NM_000358	Transforming growth factor, beta-induced, 68 kDa	2.20	0.76^a

Data are mean ± SD (n = 5).

p < 0.05 10⁻⁸ M versus 10⁻⁵ M NaI.

Genes downregulated by iodide in cultured human thyroid follicles

As shown in Table 1B, the expression of about 20 genes was decreased to less than 50% in five experiments. One of these genes is NIS (SLC5A5) (27). The gene showing the greatest decrease in expression was stearoyl-CoA desaturase (delta-9-desaturase), a regulatory enzyme that stimulates de novo synthesis of oleate and palmitoleate (28). Further, the mRNA expression levels of fatty acid desaturase 2, genes regulating cholesterol metabolism (low-density lipoprotein receptor, 7-dehydrocholesterol reductase), and genes involved in energy production (P450 [cytochrome] oxidoreductase, creatine kinase, and ATP synthase [ATP5G1]) were decreased.

Iodide-increased mRNA expression of chemokines

Among a number of chemokines, mRNAs for CX3CL1, CCL2 (monocyte chemoattractant protein-1 [MCP-1]), CXCL1, CXCL5, CXCL2, and CXCL12 were moderately expressed in cultured thyroid follicles (Fig. 1A). Further, slight but significant expression of mRNAs for CCL3 (macrophage inflammatory protein; MIP-1α), CCL4 (MIP-1β), CCL5 (regulated on activation, normal T cell expressed and secreted; RANTES), CCL17, CXCL8 (IL-8), CXCL9 (Mig), CXCL10 (IP-10), CXCL14, and CXCL17 was detected (13).

When human thyroid follicles were cultured in medium containing high iodide (10⁻⁵ M), the level of expression of mRNA for CXCL1 (growth related oncogene; GRO-α) (16) and CXCL2 (GRO-β) was increased 1.72 ± 0.79 (mean ± SD, n = 5, p > 0.05) and 1.75 ± 0.33-fold (mean ± SD, n = 5, p < 0.05), respectively (Table 3). Further, iodide significantly increased the mRNA expression level of CCL2 1.86 ± 0.68-fold (mean ± SD, n = 5, p < 0.05). In three of the five experiments, the expression was increased more than twofold. Histological examination revealed that thyroid follicles in which CCL2 mRNA expression exceeded twofold in the presence of iodide (10⁻⁵ M) were obtained from Graves' thyroid gland that had been moderately infiltrated with immunocompetent cells (data not shown). Further, mRNA expression levels of CXCL8 and CXCL14 were increased more than twofold in all experiments (n = 5) (p < 0.05). In contrast, TSH had no significant effect on these genes, as reported in the GEO (accession no. GSE12244).

Table 3.

Effect of Iodine on mRNA Expression Levels of Chemokines in Cultured Human Thyroid Follicles

Gene name	Synonyms	Major target cell showing chemotaxis	Receptor	Mean	SD
CC chemokines
CCL1	I-309	Monocytes, T cells	CCR8	1.05	0.47
CCL2	MCP-1	Momocytes, T cells, basophils, NK cells, progenitors	CCR2	1.86	0.68^a
CCL3	MIP-1α	Monocytes, T cells, NK cells, basophils, eosinophils, dendritic cells, hematopoietic progenitors	CCR1,CCR5	1.13	0.47
CCL3L3	LD78	Monocytes, T cells, NK cells, basophils, eosinophils, dendritic cells, hematopoietic progenitors	CCR1,CCR5	1.16	0.47
CCL4	MIP-1β	Monocytes, T cells, dendritic cells, NK cells, progenitors	CCR5	0.97	0.2
CCL5	RANTES	T cells, eosinophils, basophilis, NK cells, dendritic cells	CCR1, CCR3, CCR5	1.13	0.64
CCL7	MCP-3	Monocytes, T cells, eosinophils, basophils, NK cells, dendritic cells	CCR1, CCR2, CCR3	0.86	0.32
CCL8	MCP-2	Monocytes, T cells, eosinophils, basophils, NK cells	CCR1, CCR2, CCR3,	0.96	0.58
CCL11	Eotaxin	Eosonophils, T cells	CCR3	0.98	0.5
CCL13	MCP-4	Monocytes, T cells, eosinophils	CCR1, CCR2, CCR3	1.04	0.28
CCL15	HCC-2	Monocytes, T cells, eosinophils	CCR1, CCR3	1.07	0.43
CCL15				1.57	0.83
CCL16	LEC	T cells, neutrophils	CCR1	1.20	0.65
CCL17	TARC	T cells	CCR4	1.11	0.39
CCL18	PARC	Naive T cells	Unknown	1.13	0.14
CCL19	ELC	T cells, B cells, dendritic cells, activated NK cells	CCR7	1.28	0.25
CCL20	LARC	T cells, B cells	CCR6	0.96	0.15
CCL21	SLC	T cells, B cells, dendritic cells, activated NK cells	CCR7	0.97	0.24
CCL22	MDC	Macrophage progenitors	CCR4	0.76	0.39
CCL23	MIP-3	T cells, eosinophils	CCR1	1.05	0.63
CCL24	Eotaxin-2	Dendritic cells, osteoclasts	CCR3	1.00	0.02
CCL25	TECK	Effector Th2 cells	CCR9	1.12	0.48
CCL26	Eotaxin-3	Memory T cells, B cells, immature thymocytes	CCR3	1.05	0.53
CCL27	CTACK	Eosinophils, T cells	CCR10	1.04	0.17
CCL28	MEC	CLA+ T cells	CCR10, CCR3	1.37	0.34
CXCL chemokines
CXCL1	GRO-α	Neutrophils, endothelial cells	CXCR2, CXCR1	1.72	0.79
CXCL2	GRO-β	Neutrophils, endothelial cells	CXCR2	1.75	0.33^a
CXCL3	GRO-γ	Neutrophils	CXCR2	1.29	0.33
CXCL4	PF4	Fibroblasts, endothelial cells	CXCR3B	0.99	0.26
CXCL4L1	PF4V1		Unknown	1.04	0.5
CXCL5	ENA-78	Neutrophils	CXCR2	1.07	0.41
CXCL6	GCP-2	Neutrophils	CXCR1, CXCR2	0.87	0.45
CXCL7	NAP-2/PPBP	Fibroblasts	Unknown	0.95	0.51
CXCL8	IL-8	Neutrophils, T cells, basophils, endothelial cells	CXCR1, CXCR2	2.16	0.49^a
CXCL9	Mig	T cells, progenitors	CXCR3, CXCR3B	1.29	0.67
CXCL10	IP-10	T cells	CXCR3, CXCR3B	1.30	0.68
CXCL11	I-TAC	T cells	CXCR3, CXCR3B, CXCR7	1.29	0.47
CXCL12	SDF-1α/β	Monocytes, B cells, hematopoietic progenitors, nonhematopoietic cells	CXCR4, CXCR7	1.09	0.19
				1.19	0.35
CXCL13	BLC/BCA-1	T cells, naive B cells	CXCR5	1.22	0.68
CXCL14	BRAK		Unknown	2.27	0.23^a
CXCL16	SRPSOX	T cells, NK cells	CXCR6	1.07	0.12
C chemokines
XCL1	Lymphotactin-α	B cells, T cells, NK cells, neutrophils	XCR1	1.11	0.62
XCL2	Lymphotactin-β	B cells, T cells, NK cells, neutrophils	XCR1	1.15	0.68
CX3C chemokines
CX3CL1	Fraktalkine	Effector T cells	CX3CR1	0.95	0.21

Data are mean ± SD (n = 5).

p < 0.05 10⁻⁸ M versus 10⁻⁵ M NaI.

MCP-1, monocyte chemoattractant protein-1; NK, natural killer; MIP, macrophage inflammatory protein; RANTES, regulated on activation, normal T cell expressed and secreted; GRO, growth related oncogene.

Among 20 chemokine receptors, CXCR7, CXCR4, and CXCR3 were most abundantly expressed in thyroid follicles (Fig. 1B). However, iodide did not significantly modulate their expression (data not shown). Iodide had no significant effect on the mRNA expression of cytokines such as IL-1, tumor necrosis factor-alpha, and interferon-gamma (data not shown).

Expression levels of various genes determined by RT-PCR

To confirm that iodide induces several chemokines, real-time PCR was performed (Figs. 2 and 3). As has been well documented (27), the expression of NIS mRNA gradually decreased to less than 20% in the presence of a high iodide concentration (10⁻⁵ M) (Fig. 2A). As expected, iodide (10⁻⁵ M) increased the expression of TGFβ1, and consequently TGFBI, more than twofold by 48 hours (Fig. 2B, C). Expression of ICAM-1 mRNA was also increased more than twofold at 24 hours (Fig. 2D). Further, the mRNA expression levels of CCL2, CXCL8 (IL-8), and CXCL14 increased time dependently more than twofold (Fig. 2E–G).

FIG. 2.

Real-time polymerase chain reaction (PCR) analyses of the effect of iodide on the mRNA expression levels of various genes. Thyroid follicles were precultured for 5 days and then cultured in medium containing bovine TSH (30 μU/mL) and a low (10⁻⁸ M) or high (10⁻⁵ M) iodide concentration. After an additional 6 to 72 hours of culture, total RNA was extracted, and the mRNA expression levels were analyzed by real-time PCR, as described in the Materials and Methods section. Data are mean ± SD of three samples. (A) NIS, (B) TGFβ1, (C) TGFBI, (D) ICAM1, (E) CCL2, (F) CXCL8, (G) CXCL14, (H) ANXA1, (I) IL-6 interleukin-6.

FIG. 3.

Dose–response effect of iodide on expression of mRNAs for various genes. Thyroid follicles were precultured for 5 days and then cultured in medium containing bovine TSH (30 μU/mL) and various concentrations of iodide (10⁻⁸ to 10⁻² M). After an additional 72 hours of culture, total RNA was extracted, and the mRNA expression levels were analyzed by real-time PCR, as described in the Materials and Methods section. Data are mean ± SD of three samples. (A) ICAM1, (B) CCL2, (C) CXCL8, (D) CXCL14.

The mRNA expression level of ANXA1 was also increased in a time-dependent manner (Fig. 2H). In contrast, there was no significant change in the expression level of IL-6 mRNA (Fig. 2I). These data were all consistent with those obtained from microarray analyses of thyroid follicles that had been cultured in high iodide medium (10⁻⁵ M) for 48 hours.

In another experiment, total RNAs were extracted from thyroid follicles that had been cultured in medium containing bovine TSH (30 μU/mL) and various concentrations of iodide (10⁻⁸ to 10⁻² M) for 48 hours. RT-PCR analyses revealed that iodide dose dependently increased the mRNA expression of ICAM1(Fig. 3A), CCL2 (Fig. 3B), CXCL8 (Fig. 3C), and CXCL14 (Fig. 3D) at 10⁻⁶ to 10⁻⁴ M. However, at supraphysiological concentrations (10⁻³ to 10⁻² M), the stimulatory effects of iodide were obscured, probably due to the cytotoxic effects of iodide.

Immunohistochemical studies of cultured thyroid follicles and thyroid glands

To confirm whether the observed chemokines are expressed in the follicular epithelium, formalin-fixed, paraffin-embedded thyroid tissues obtained from five patients with Graves' disease by subtotal thyroidectomy were subjected to immunohistochemistry after antigen retrieval. In all the Graves' thyroids, CCL2, CXCL8, and CXCL14 were demonstrated at various staining intensities in the follicular epithelium (Fig. 4A–C). Strong immunoreactivity was occasionally observed in follicles where lymphocytic infiltration was evident. There was a tendency for immunostaining to be stronger in epithelium close to lymphocyte infiltration. Further, CXCR7 was also demonstrated in the thyroid follicular cells (Fig. 4D).

FIG. 4.

Immunohistochemistry of chemokines in Graves' thyroid gland. Thyroid tissues were stained with antibodies against CCL2 (A), CXCL8 (B), CXCL14 (C), and CXCR7 (D) after antigen retrieval, as described in the Materials and Methods section. Original magnification × 200. Color images available online at www.liebertonline.com/thy.

Discussion

By culturing human thyroid follicles obtained from patients with Graves' disease, we confirmed our previous observation that the expression of CXCL8 (IL-8, MDNCF) (29), a chemokine that attracts mostly neutrophils in addition to T cells and endothelial cells, was increased more than twofold when cultured in the presence of a high iodide concentration (10⁻⁵ M). Although this cytokine (previously called MDNCF) does not affect TSH-induced thyroid function per se (30), it would attract immunocompetent cells into thyroid follicles, leading to thyroid dysfunction if immune cells have infiltrated the thyroid gland. Because lymphocytes cannot incorporate ¹²⁵I (30) and iodide at high concentrations (10⁻⁶ to 10⁻⁴M) can stimulate H₂O₂ generation in thyroid slices (31), reactive oxygen species (ROS) induced by H₂O₂ in thyrocytes may be at least partly involved in IL-8 production (32).

We also confirmed our previous observation that iodide increased the expression level of TGF-β1 mRNA nearly twofold (11,12) and demonstrated that TGFBI mRNA expression also increased more than twofold (Table 1A, Fig. 2B, C). Because TGFβ1 is known to directly stimulate the expression of CCL-2 (MCP-1) in cultured human thyroid follicular cells (33), this chemokine produced by infiltrated inflammatory cells and/or thyroid follicular cells may play a significant role in thyroid immune/inflammatory responses. Further, we have confirmed that iodide in a time- and concentration-dependent manner increases the expression of ICAM1, a molecule that promotes cell-to-cell interactions and enhances the inflammatory process, probably via ROS produced in thyrocytes (20).

We demonstrated for the first time that iodide per se nearly doubled the expression level of CCL2 (MCP-1) mRNA (1.86 ± 0.68, mean ± SD, n = 5) in cultured human thyroid follicles (p < 0.05). The expression was increased more than twofold when thyroid follicles were obtained from thyroid glands that had been infiltrated by immunocompetent cells. CCL2 or MCP-1 attracts monocytes, memory T lymphocytes, and natural killer cells in vitro (34) and may regulate T-cell differentiation (35). Indeed, CXCL8 (IL-8) and CCL2 (MCP-1) are essentially involved in inflammatory and immune reactions (32).

Interestingly, breast–kidney-expressed chemokine (BRAK/CXCL14) was also increased by iodide at high concentration (p < 0.05). CXCL14 is an ill-defined chemokine with an unknown receptor expressed in the epithelial tissues and is chemotactic for dendritic cell precursors (36). CXCL14 is also selectively localized in skin fibroblasts and involved in the generation of tissue macrophages by recruiting extravasated monocytes to fibroblasts, thus promoting macrophage development (37). This chemokine is a potent inhibitor of in vitro angiogenesis (38), thus contributing at least partly to the iodine-induced decrease in thyroid blood flow.

Compared with chemokines, iodide elicited no significant effect on the expression of mRNAs for chemokine receptors. The most highly expressed chemokine receptors were CXCR7 and CXCR4. CXCR7 is highly expressed in monocytes and mature B cells and has been demonstrated in thyroid follicular cells (Fig. 4D). CXCR7 binds with high affinity to CXCL11 and CXCL12, although its chemokine signaling has not been demonstrated (39). The expression levels and regulatory mechanism of chemokine receptors in the thyroid gland remain to be clarified.

It should be pointed out that iodide stimulates not only chemokines but also several genes that are involved in immunological function. For example, ANXA1 plays a pivotal role in the innate and adaptive immune system (24). Early growth response gene 1 is also involved in the regulation of the immune response by activating TGF-β1 in kidney and ICAM-1 in B lymphocytes (25). Further, guanylate-binding protein 1 is selectively induced by inflammatory cytokines and is a marker of endothelial cell activation in inflammatory disease (26). It remains to be elucidated in future studies whether these genes are involved in iodine-induced hypothyroidism.

It should be also pointed out that our present data have possible limitations applicable to normal thyroid gland, because our experiments were performed in thyroid follicles obtained from Graves' thyroid gland that are minimally to moderately infiltrated with immune cells.

In summary, we have demonstrated that iodide significantly increases the expression levels of mRNA for CCL2, CXCL8, and CXCL14 in a time- and concentration-dependent manner in cultured human thyroid follicles. Overexpression of these chemokines together with ICAM1 probably plays a significant role in attracting immune cells into the thyroid gland and influences the course of autoimmune thyroid disease. Because these immune cells are capable of producing cytokines (IL-1, tumor necrosis factor-alpha, interferon-gamma) that synergistically inhibit thyroid hormonogenesis (40), thyroid dysfunction may develop in certain patients with autoimmune thyroid diseases. As stated by Wolff, excess iodide inhibits the thyroid through multiple mechanisms (41). Our present in vitro findings suggest that excess iodide may inhibit thyroid function through not only a biochemical but also an immunological mechanism, via stimulation of chemokines such as CCL2, CXCL8, and CCL14, leading to recruitment of immunocompetent cells into the thyroid gland and aggravation of thyroid function.

Footnotes

Disclosure Statement

The authors declare that no competing financial interests exist.

References

Markou

, Georgopoulos

, Kyriazopoulou

, Vagenakis

. 2001. Iodine-induced hypothyroidism. Thyroid, 11:501–510.

Papanastasiou

, Vatalas

, Koutras

, Mastorakos

. 2007. Thyroid autoimmunity in the current iodine environment. Thyroid, 17:729–739.

Sato

, Ohmori

, Shiratori

, Yamazaki

, Yamada

, Kimura

, Takano

. 2007. Povidone iodine-induced overt hypothyroidism in a patient with prolonged habitual gargling: urinary excretion of iodide after gargling in normal subjects. Intern Med, 46:391–395.

Braverman

, Woeber

, Ingbar

. 1969. Induction of myxedema by iodide in patients euthyroid after radioiodin or surgical treatment of diffuse toxic goiter. N Engl J Med, 281:816–821.

Yamada

, Sato

, Hara

, Tochimoto

, Takagi

, Hizuka

, Takano

. 2002. Propylthiouracil-induced lupus-like syndrome developing in a Graves' patient with a sibling with systemic lupus erythematosus. Intern Med, 41:1204–1208.

Yamazaki

, Sato

, Shizume

, Kanaji

, Ito

, Obara

, Nakagawa

, Koizumi

, Nishimura

. 1995. Potent thyrotropic activity of human chorionic gonadotropin variants in terms of ¹²⁵I incorporation and de novo—synthesized thyroid hormone release in human thyroid follicles. J Clin Endocrinol Metab, 80:473–479.

Sato

, Yamazaki

, Yamada

. 2008. DNA microarray analysis of effects of TSH iodide, cytokines, therapeutic agents on gene expression in cultured human thyroid follicles. Handwerger

, Bruce

. Genomics in Endocrinology: DNA Microarray Analysis in Endocrine Health and Diseases (Contemporary Endocrinology) Humana Press: Totowa, NJ, 207–218.

Rotondi

, Chiovato

, Romagnani

, Serio

, Romagnani

. 2007. Role of chemokines in endocrine autoimmune diseases. Endocr Rev, 28:492–520.

Kimura

, Caturegli

. 2007. Chemokine orchestration of autoimmune thyroiditis. Thyroid, 17:1005–1011.

10.

Balestrieri

, Balestrieri

, Mancini

, Napoli

. 2008. Understanding the immunoangiostatic CXC chemokine network. Cardiovasc Res, 78:250–256.

11.

Yamazaki

, Yamada

, Kanaji

, Yanagisawa

, Kato

, Takano

, Obara

, Sato

. 2003. Genes regulated by TSH and iodide in cultured human thyroid follicles: analysis by cDNA microarray. Thyroid, 13:149–158.

12.

Yamada

, Yamazaki

, Takano

, Obara

, Sato

. 2006. Iodide inhibits vascular endothelial growth factor (VEGF)-A expression in cultured human thyroid follicles: a microarray search for effects of TSH and iodide on angiogenesis factors. Thyroid, 16:545–554.

13.

Kemp

, Metcalfe

, Smith

, Woodroofe

, Watson

, Weetman

. 2003. Detection and localization of chemokine gene expression in autoimmune thyroid disease. Clin Endocrinol (Oxf), 59:207–213.

14.

Ashhab

, Dominguez

, Sospedra

, Roura-Mir

, Lucas-Martin

, Pujol-Borrell

. 1999. A one-tube polymerase chain reaction protocol demonstrates CC chemokine overexpression in Graves' disease glands. J Clin Endocrinol Metab, 84:2873–2882.

15.

Aust

, Steinert

, Boltze

, Kiessling

, Simchen

. 2001. GRO-alpha in normal and pathological thyroid tissues and its regulation in thyroid-derived cells. J Endocrinol, 170:513–520.

16.

Ferrer-Francesch

, Caro

, Alcalde

, Armengol

, Ashhab

, Lucas-Martin

, Martinez-Caceres

, Juan

, Pujol-Borrell

. 2006. One-tube-PCR technique for CCL2, CCL3, CCL4 and CCL5 applied to fine needle aspiration biopsies shows different profiles in autoimmune and non-autoimmune thyroid disorders. J Endocrinol Invest, 29:342–349.

17.

Nakajima

, Yamazaki

, Yamada

, Kanaji

, Kosaka

, Sato

, Takano

. 2001. Amiodarone stimulates interleukin-6 production in cultured human thyrocytes, exerting cytotoxic effects on thyroid follicles in suspension culture. Thyroid, 11:101–109.

18.

Yamazaki

, Suzuki

, Yamada

, Takeshita

, Matsumoto

, Mitsuhashi

, Obara

, Takano

, Sato

. 2007. Suppression of iodide uptake and thyroid hormone synthesis with stimulation of type I interferon system by double-stranded RNA (dsRNA) in cultured human thyroid follicles. Endocrinology, 48:3226–3235.

19.

Sharma

, Alegria

, Talor

, Rose

, Caturegli

, Burek

. 2005. Iodine and IFN-gamma synergistically enhance intercellular adhesion molecule 1 expression on NOD.H2h4 mouse thyrocytes. J Immunol, 174:7740–7745.

20.

Sharma

, Traore

, Trush

, Rose

, Burek

. 2008. Intracellular adhesion molecule-1 up-regulation on thyrocytes by iodine of non-obese diabetic.H2(h4) mice is reactive oxygen species-dependent. Clin Exp Immunol, 152:13–20.

21.

Kostic

, Toffoletto

, Fontanini

, Moretti

, Cesselli

, Beltrami

, Ambesi Impiombato

, Curcio

. 2009. Influence of iodide excess and interferon-gamma on human primary thyroid cell proliferation, thyroglobulin secretion, and intracellular adhesion molecule-1 and human leukocyte antigen-DR expression. Thyroid, 19:283–291.

22.

Yuasa

, Eggo

, Meinkoth

, Dillmann

, Burrow

. 1992. Iodide induces transforming growth factor beta 1 (TGF-beta 1) mRNA in sheep thyroid cells. Thyroid, 2:141–145.

23.

Cowin

, Bidey

. 1995. Porcine thyroid follicular cells in monolayer culture activate the iodide-responsive precursor form of transforming growth factor-beta 1. J Endocrinol, 144:67–73.

24.

D'Acquisto

, Perretti

, Flower

. 2008. Annexin-A1: a pivotal regulator of the innate and adaptive immune systems. Br J Pharmacol, 155:152–169.

25.

McMahon

, Monroe

. 1996. The role of early growth response gene 1 (egr-1) in regulation of the immune response. J Leukoc Biol, 60:159–166.

26.

Lubeseder-Martellato

, Guenzi

, Jörg

, Töpolt

, Naschberger

, Kremmer

, Zietz

, Tschachler

, Hutzler

, Schwemmle

, Matzen

, Grimm

, Ensoli

, Stürzl

. 2002. Guanylate-binding protein-1 expression is selectively induced by inflammatory cytokines and is an activation marker of endothelial cells during inflammatory diseases. Am J Pathol, 161:1749–1759.

27.

Dohán

, De la

Vieja A

, Paroder

, Riedel

, Artani

, Reed

, Ginter

, Carrasco

. 2003. The sodium/iodide symporter (NIS): characterization, regulation, and medical significance. Endocr Rev, 24:48–77.

28.

Ntambi

, Miyazaki

, Dobrzyn

. 2004. Regulation of stearoyl-CoA desaturase expression. Lipids, 39:1061–1065.

29.

Hoffmann

, Dittrich-Breiholz

, Holtmann

, Kracht

. 2002. Multiple control of interleukin-8 gene expression. J Leukoc Biol, 72:847–855.

30.

Sato

, Yamazaki

, Shizume

, Yamakawa

, Satoh

, Kanaji

Jr. , Obara

, Fujimoto

, Aiba

, Kakiuchi

, Kanaji

. 1993. Pathogenesis of autoimmune hypothyroidism induced by lymphokine-activated killer (LAK) cell therapy: in vitro inhibition of human thyroid function by interleukin-2 in the presence of autologous intrathyroidal lymphocytes. Thyroid, 3:179–188.

31.

Corvilain

, Collyn

, van Sande

, Dumont

. 2000. Stimulation by iodide of H₂O₂ generation in thyroid slices from several species. Am J Physiol Endocrinol Metab, 278:E692–E699.

32.

Mukaida

, Harada

, Matsushima

. 1998. Interleukin-8 (IL-8) and monocyte chemotactic and activating factor (MCAF/MCP-1), chemokines essentially involved in inflammatory and immune reactions. Cytokine Growth Factor Rev, 9:9–23.

33.

Matsumura

, Banba

, Motohashi

, Hattori

. 1999. Interleukin-6 and transforming growth factor-beta regulate the expression of monocyte chemoattractant protein-1 and colony-stimulating factors in human thyroid follicular cells. Life Sci, 65:129–135.

34.

Menten

, Wuyts

, Van

Damme J

. 2002. Macrophage inflammatory protein-1. Cytokine Growth Factor Rev, 13:455–481.

35.

Luther

, Cyster

. 2001. Chemokines as regulators of T cell differentiation. Nat Immunol, 2:102–107.

36.

Gharaee-Kermani

, Denholm

, Phan

. 1996. Costimulation of fibroblast collagen and transforming growth factor beta1 gene expression by monocyte chemoattractant protein-1 via specific receptors. J Biol Chem, 271:17779–17784.

37.

Kurth

, Willimann

, Schaerli

, Hunziker

, Clark-Lewis

, Moser

. 2001. Monocyte selectivity and tissue localization suggests a role for breast and kidney-expressed chemokine (BRAK) in macrophage development. J Exp Med, 194:855–861.

38.

Shellenberger

, Wang

, Gujrati

, Jayakumar

, Strieter

, Burdick

, Ioannides

, Efferson

, El-Naggar

, Roberts

, Clayman

, Frederick

. 2004. BRAK/CXCL14 is a potent inhibitor of angiogenesis and a chemotactic factor for immature dendritic cells. Cancer Res, 64:8262–8270.

39.

Thelen

, Thelen

. 2008. CXCR7, CXCR4 and CXCL12: an eccentric trio? J Neuroimmunol, 198:9–13.

40.

Sato

, Satoh

, Shizume

, Ozawa

, Han

, Imamura

, Tsushima

, Kanaji

Jr. , Ito

, Obara

, Fujimoto

, Kanaji

. 1990. Inhibition of ¹²⁵I organification and thyroid hormone release by interleukin-1, tumor necrosis factor-α, interferon-γ in human thyrocytes in suspension culture. J Clin Endocrinol Metab, 70:1735–1743.

41.

Wolff

. 1989. Excess iodide inhibits the thyroid by multiple mechanisms. Adv Exp Med Biol, 261:211–244.