| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Medical Department P, Endocrine Division (A.K.R., U.F.-R.); the Institute for Inflammation Research, (M.D., K.B.); and the Department of Endocrine Surgery (M.B.-T.), Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
Address all correspondence and requests for reprints to: Å. Krogh Rasmussen, M.D., Medical Department P2131, Endocrine Division, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen Ø, Denmark.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
High affinity IL-1 receptors have been demonstrated on thyroid cells in (7, 8). However, it remains to be clarified whether all functions of IL-1ß on the thyrocytes are mediated through these receptors and through which secondary pathways. For example, IL-1ß induced the production of nitric oxide (NO) in human thyrocytes (9, 9a), but this appeared to be unrelated to previously described effects of IL-1ß on the cells (9). NO exerts many of its effects via activation of guanylate cyclase and an increase in cGMP.
The aims of the present study were, therefore, to determine whether the cGMP pathway was involved in the action of IL-1 on human thyroid cells, and whether the naturally occurring IL-1 receptor antagonist (IL-1ra) could block all IL-1ß-induced functional changes in cultured monolayers of human thyrocytes. We also investigated whether these IL-1ß-induced effects were reversible.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TNF; 40 IU/ng) was provided by Dr. G. R. Adolf
(Ernst Boehringer Institute, Vienna, Austria). The biological activity
of IL-1ß was confirmed by EL4 bioassay (10), and the activity of
recombinant TNF was confirmed by an L-M fibroblast cytotoxicity
assay (11). The activities were calibrated by the use of international
interim reference preparations (National Institute of Biological
Standards and Controls, Potters Bar, UK). IL-1ra was provided by Daniel
Tracey (Upjohn Co., Kalamazoo, MI). The cytokine preparations contained
less than 0.2 fg endotoxin/ng cytokine, as measured by a chromogenic
Limulus amebocyte lysate assay (Struers, Rodovre, Denmark).
Thyroid cell cultures
Human thyroid tissue specimens were obtained from
paraadenomatous tissue from 10 patients undergoing thyroidectomies.
Informed consent was obtained from all subjects. Secondary cell
cultures in monolayers were prepared as previously described (12). The
cells were cultured in Ham’s F-12 medium modified by Coon’s medium
supplemented with six hormones and growth factors [bovine TSH, 1
U/liter (Rhone-Poulenc-Rorer, Armour, Collegeville, PA); human insulin,
10 mg/liter (Boehringer Mannheim, Mannheim, Germany); somatostatin, 10
µg/liter (Novabiochem, Laufelfingen, Switzerland); human transferrin,
6 mg/liter (Boehringer Mannheim); hydrocortisone, 10-8
M (Calbiochem, La Jolla, CA); and
glycyl-histidyl-lysine acetate, 10 µg/liter (Calbiochem)] and in the
presence of L-glutamine (2 mM; Biological
Industries, Beth Haemek, Israel), antibiotics (100 kU/liter penicillin
and 100 mg/liter streptomycin; Biological Industries), as well as
nonessential amino acids (Biological Industries) and 5% FCS. TSH
stimulates thyroid function, whereas the other five hormones/growth
factors did not (13). The purities of the cell preparations were high,
as immunocytochemical analysis of the cells in secondary cultures
revealed 99% Tg-positive cells (12).
Cell functions
The functions of the cells were evaluated by the release of Tg,
cAMP, cGMP, and IL-6 into the medium. The medium was stored at -20 C
until measurement. Tg and IL-6 were measured by enzyme-linked
immunosorbent assays (12, 14), and cAMP was determined by a competitive
protein binding assay (15) in the presence of the phosphodiesterase
inhibitor, 3-isobutyl-L-methylxanthine (0.5
mM). cGMP was measured by a 125I RIA kit
(Advanced Magnetics, Cambridge, MA).
To correct for the variation in cell numbers in the chambers, the
concentrations of Tg, cAMP, cGMP, and IL-6 were related to the DNA
contents of each chamber, as measured by a modification of the
previously described diphenylamine method (16). IL-1ra production from
the thyrocytes was measured by a double sandwich enzyme-linked
immunosorbent assay using monospecific polyclonal rabbit antibodies
against purified recombinant IL-1ra, as previously described for
IL-1 (17). NO was quantified as previously described (9).
Experimental procedure
In all experiments the cells were deprived of TSH for 72 h
before the test. The cells were exposed to rIL-1ß with an activity
between 1–105 IU/liter (0.0002–20 µg/liter) or to
IL-1ß (20 µg/liter = 105 IU/liter) in combination
with TNF (25 µg/liter = 106 IU/liter) for 72
h. IL-1ra (2.5–500 µg/liter) was added together with the cytokines.
Thyroid cells were also preexposed to IL-1ß (2 or 20 µg/liter) for
48 h, followed by 24 h with IL-1ß in combination with
IL-1ra (250–500 µg/liter). The influence of sodium nitroprusside
(SNP; 1–10-100 µM), a chemical generator of
NO, on cGMP release in the presence or absence of TSH was studied, and
it was investigated whether the competitive inhibitor of NO synthase,
L-NG-mono-methyl-arginine (L-NMMA;
0.1 mM), could modulate the IL-1ß-induced effect on cGMP
release from the cells. Both SNP (S 0501, Sigma Chemical Co., St.
Louis, MO) and L-NMMA (Sigma M 7033) were dissolved in medium
immediately before use.
Statistics
The experiments were carried out in two to eight different
cultures in at least triplicate. Results were expressed as the mean and
SD and analyzed parametrically using unpaired Student’s
t test (two-sample analysis) or one-way ANOVA, comparing the
responses at different concentrations of the same cytokine (unbalanced
design; Statgraphics version 2.6, STSC, Rockville, MD).
P < 0.05 was considered statistically significant.
Analogous results and conclusions were found using nonparametric
analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and Table 1; P < 0.0001 (n
= 9–38 depending on the IL-1ß level) and P < 0.0001
(n = 9–38), respectively, by one-way ANOVA], whereas TNF (25
µg/liter) enhanced the IL-1ß-induced inhibition of both Tg
(P = 0.003; n = 13/35) and cAMP (P
= 0.02; n = 14/35; Table 1). Tg and cAMP release from the
thyrocytes was significantly increased and almost restored after
removal of IL-1ß (Fig. 1, A and B); at the same time, IL-1ß-induced
IL-6 production decreased significantly after removal of IL-1ß
(Fig. 1C). IL-1ß induced a dose-dependent release of cGMP
(P < 0.001, by one-way ANOVA; n = 6–23), which
was significantly reduced by the competitive inhibitor of NO synthase,
L-NMMA (Fig. 2). In contrast, TNF (2.5 and 25
µg/liter) was unable to induce cGMP production [mean ±
SD, 520 ± 50 and 730 ± 100 fmol/µg DNA,
respectively (n = 6), compared to the control (1,820 ± 1,420
fmol/µg DNA; n = 21)]. TNF was also unable to enhance
IL-1ß-induced cGMP release [IL-1ß (20 µg/liter) alone induced
12,540 ± 9,830 fmol cGMP/µg DNA, and IL-1ß (20 µg/liter)
together with TNF (25 µg/liter) induced 8,500 ± 4,870 fmol
cGMP/µg DNA; P > 0.05; n = 17–23]. IL-6
production by thyroid cells was stimulated dose dependently by IL-1ß
(Fig. 3; P < 0.0001, by one-way ANOVA;
n = 9–29). TNF (25 µg/liter) did not enhance this IL-1ß
(20 µg/liter)-induced IL-6 production (P > 0.05;
n = 14/29).
|
|
|
|
|
and Table 1A) and cAMP (Fig. 3B and Table 1B).
However, IL-1ra (250 µg/liter) could not totally eliminate the
inhibitory influence of IL-1ß at the highest concentration. IL-1ra at
low and medium concentrations (2.5 and 25 µg/liter) was unable to
protect against the inhibition of Tg and cAMP (Tables 1A and 1B). The
same pattern was found in IL-1-exposed thyroid cells (data not
shown). IL-1ra (250 µg/liter) significantly reduced the inhibitory
influence on Tg release, but not cAMP induced by IL-1ß (20
µg/liter) combined with TNF (25 µg/liter; Tables 1A and 1B).
IL-1ra (250 µg/liter) by itself did not modify the release of Tg and
cGMP, but inhibited the release of cAMP (P = 0.003;
n = 34/38; Figs. 2 and 3 and Table 1). IL-1ra added after
preexposure to IL-1ß for 48 h reduced the protective ability of
IL-1ra against the inhibitory influence on Tg and cAMP production
(Table 2). IL-1ra (125 µg/liter) further reduced
IL-1ß (2 and 20 µg/liter)-induced cGMP release, which also was
significantly reduced by L-NMMA (0.1 mM; Fig. 2).
|
IL-1ra (250 µg/liter) added together with IL-1ß significantly
diminished IL-1ß-induced IL-6 production (Fig. 3C), but not the IL-6
production induced by IL-1ß (20 µg/liter) in combination with
TNF (25 µg/liter). IL-1ra (25 and 2.5 µg/liter) alone was unable
to reduce the IL-6 production.
Influence of SNP
SNP (1–10-100 µM) enhanced the
accumulation of cGMP dose-dependently in both the presence and absence
of TSH (P < 0.001 and P < 0.0001,
respectively, by one-way ANOVA; n = 12). At 100 µM
SNP, TSH-stimulated cGMP release was 13,267 ± 7,829 fmol/µg
DNA, and in the absence of TSH, it was 13,133 ± 7,803 fmol/µg
DNA compared to 812 ± 133 fmol/µg DNA in the control. TSH by
itself had no influence on cGMP release. Furthermore, SNP generated
dose-dependent NO production independent of TSH. At 100
µM SNP, TSH-stimulated NO generation was 62.5 ±
14.1 µmol/liter; in the absence of TSH it was 58.5 ± 15.0
µmol/liter compared to 10.2 ± 0.4 µmol/liter in the control
(P < 0.0001; n = 6). NO generation induced by SNP
was significantly greater than cytokine-induced NO generation (data not
shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-1ß and consists of a
single polypeptide chain of 17 kDa (18). It is produced by activated
macrophages and monocytes, among other cells, and binds selectively to
IL-1 receptors without causing internalization of the receptor or
activating the target cell. IL-1ra competes with IL-1 and IL-1ß
for occupancy of the IL-1 receptors, and it selectively blocks the
biological activities of IL-1 and IL-1ß in several in
vitro and in vivo models (reviewed in Refs. 19 and 20).
Blockade of the binding of IL-1 to the thyroid cell could be a way to
protect the cells from the effects of IL-1. However, perhaps not all of
the effects of the IL-1 agonists IL-1/ß are mediated through the
presently described configurations of IL-1 receptors and accessory
molecules (21). The data presented here confirm previous findings that moderate levels of IL-1ß decrease TSH-stimulated human thyroid cell release of Tg and cAMP (3, 4, 5) and increase IL-6 production in human thyroid cells (6). All of these effects were counteracted by IL-1ra. The effects were reversible, indicating a noncytotoxic mechanism of the IL-1ß influence. Furthermore, IL-1ß stimulated the release of cGMP from the cells, an effect that was significantly decreased by L-NMMA and counteracted by IL-1ra. The effect of IL-1ß on IL-6 production by the thyrocytes was reduced in the presence of IL-1ra, in accordance with the findings of Granowitz et al. (22). The diminishing effect of IL-1ra on the effects of IL-1ß on human thyroid cells indicated that the described IL-1ß influences were mediated through IL-1 receptors on the thyrocytes. The demonstrated SNP-induced TSH-independent stimulation of cGMP release was also found by Millatt and co-workers (23), and also SNP-induced NO generation was independent of TSH, indicating that the NO-cGMP pathway is not involved after TSH receptor activation.
IL-1ra appeared to interfere with early processes initiated by IL-1ß,
as the addition of IL-1ra after exposure of the cells to IL-1ß
diminished their ability to counteract the effects of the cytokine.
Even though IL-1ra has the same affinity for IL-1 receptor-bearing
cells as IL-1 and IL-1ß, at least a 10-fold molar excess is
usually required to block the biological activity of IL-1/ß (20).
This most likely reflects the fact that very few receptors need
activation by IL-1 and/or IL-1ß to activate the target cells.
Accordingly, a 10-fold molar excess of IL-1ra was required in this
study to inhibit the IL-1-induced response in thyrocytes. In a
different cell system of cultured rat pancreatic ß-cells, a 10- to
10,000-fold excess of IL-1ra was required to protect the cells against
the effects of IL-1ß (24, 25). A total blockade of the effects of
IL-1ß on thyrocyte function induced by IL-1ra could be obtained by
the use of a lower concentration of IL-1ß combined with a high
concentration of IL-1ra. Thus, it appears that the demonstrated IL-1ß
effects are mediated entirely via specific IL-1 receptors on the
thyrocytes.
The influence induced by IL-1ß and TNF in combination was only
sparsely counteracted by IL-1ra. This may partly be a question of the
requirement for a higher concentration of IL-1ra, but may also suggest
that at least part of the TNF effect was direct and not mediated
through TNF-induced generation of IL-1 in the thyrocytes.
Naturally produced IL-1ra is thought to be an important in vivo inhibitor of IL-1-induced functions on various target cells. IL-1ß is cytotoxic to pancreatic rodent ß-cells and suppresses insulin production and release from the cells, and IL-1ra protects the insulin-producing cells from the effect of IL-1ß (24, 25, 26, 27). Similarly, sustained administration of IL-1ra can delay or prevent the diabetogenic processes in BB rats and in multiple low dose streptozotocin-treated mice (28, 29).
In conclusion, although IL-1ß inhibited adenylate cyclase-mediated
pathways, it stimulated guanylate cyclase. Both effects could be
blocked by IL-1ra, indicating that the effects were exerted through
activation of specific IL-1 receptors on thyrocytes. IL-1 may act at
multiple intracellular sites in the thyroid. It has been shown that
some of the known effects of IL-1 on the thyroid are not mediated via
the NO-cGMP pathway (9), and it is unclear which effects induced by
IL-1 are mediated through an activation of this pathway. IL-1 has
been demonstrated to induce loss of the thyroid epithelial barrier,
measured as transepithelial resistance (30), an effect that might be
mediated via NO release. Both guanylate cyclase- and adenylate
cyclase-mediated pathways may be involved in the cellular response of
the thyrocyte in thyroid autoimmunity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipient of a fellowship award from the Alfred Benzon Foundation
(Copenhagen, Denmark).
Received August 20, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 1ß on cultured human and rat thyroid cells. Acta Endocrinol
(Copenh) 122:4:520–526
, TNF-ß and IFN-
, separately and
combined with IL-1ß, on the function of cultured human thyroid cells.
J Endocrinol 143:359–365 antibodies. Immunol Lett 30:133–140[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  S. Poncin, B. Lengele, I. M. Colin, and A.-C. Gerard Differential Interactions between Th1/Th2, Th1/Th3, and Th2/Th3 Cytokines in the Regulation of Thyroperoxidase and Dual Oxidase Expression, and of Thyroglobulin Secretion in Thyrocytes in Vitro Endocrinology, April 1, 2008; 149(4): 1534 - 1542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Flouris, G. S. Metsios, A. Z. Jamurtas, and Y. Koutedakis Sexual dimorphism in the acute effects of secondhand smoke on thyroid hormone secretion, inflammatory markers and vascular function Am J Physiol Endocrinol Metab, February 1, 2008; 294(2): E456 - E462. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mühlberg, W. Joba, C. Spitzweg, H.-D. Schworm, H.-J. Heberling, and A. E. Heufelder Interleukin-1 Receptor Antagonist Ribonucleic Acid and Protein Expression by Cultured Graves' and Normal Orbital Fibroblasts Is Differentially Modulated by Dexamethasone and Irradiation J. Clin. Endocrinol. Metab., February 1, 2000; 85(2): 734 - 742. [Abstract] [Full Text]
|
|
|
|
|
|
D. F. Sellitti and S. Q. Doi Regulation of Natriuretic Peptide Receptors by Thyrotropin in FRTL-5 Rat Thyroid Cells: Evidence for Nonguanylate Cyclase Atrial Natriuretic Factor-Binding Sites in Cells Lacking the Natriuretic Peptide Receptor C Endocrinology, March 1, 1999; 140(3): 1365 - 1374. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
; *,
P < 0.05; **, P < 0.001
compared to the control), and release of Tg, cAMP, and IL-6 from the
same human thyrocytes after removal of IL-1ß for an additional
72 h (
; *, P < 0.05; **,
P < 0.001 compared to the influence of 72-h
IL-1ß).
) and
IL-1ra (125 µg/liter; 