| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Immunology, Erasmus University Rotterdam, 3000 DR Rotterdam, The Netherlands
Address all correspondence and requests for reprints to: P. J. Simons, Lab. Ee 838, Department of Immunology, Faculty of Medicine, Erasmus University Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. E-mail: simons{at}immu.fgg.eur.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We investigated the effect of (Nycodenz density gradient) purified
splenic DC from Wistar rats on the growth rate of and thyroid hormone
secretion by Wistar thyroid follicles (collagenase dispersion) in
culture. Various numbers of DC and follicles were cocultured during
24 h. The proliferative capacity of thyrocytes was measured by
adding tritiated thymidine (3H-TdR) and bromodeoxyuridine,
the hormone secretion into the culture fluid was measured by using a
conventional T3 RIA. Furthermore, antibodies directed
against interleukin-1ß (IL-1ß), IL-6, and tumor necrosis factor-
(TNF-) were added to these cocultures to determine the role of these
cytokines in a possible DC regulation of thyrocyte growth. Cocultures
were also carried out in the presence of antimajor histocompatibility
complex-class I (MHC I), anti-MHC II, antiintercellular adhesion
molecule-1 (ICAM-1), and antilymphocyte function-associated
antigen-1 (LFA-1) antibodies to possibly interfere with
DC-thyrocyte interactions.
The addition of DC to thyroid follicles clearly inhibited their
3H-TdR uptake, particularly at a 10:1 ratio, in comparison
to follicle cultures alone, both under basal conditions and after TSH
stimulation (75 ± 7% and 49 ± 11% reduction,
respectively, n = 4). The follicle T3 secretion (after
TSH stimulation) was also suppressed by DC in this system, but to a
lesser extent (at best at an 1:1 ratio, 25 ± 7% reduction,
n = 4). The DC-induced inhibition of thyroid follicle growth was
totally abrogated after addition of anti-IL-1ß antibodies; anti-IL-6
only had effect on the DC inhibition of non-TSH-stimulated thyrocytes,
whereas anti-TNF- demonstrated no effect at all. The antibodies to
MHC and to adhesion molecules had also no effect on this DC-induced
growth inhibition. The effect of the different anti-cytokine and
anti-adhesion antibodies on the T3 secretion from thyroid
follicles was not investigated.
The clear inhibition of thyrocyte growth by splenic DC (classical antigen-presenting cells) again demonstrates the regulatory role of DC in endocrine systems. Proinflammatory cytokines such as IL-1ß and IL-6 are important mediators in this regulation. The here shown dual role of DC represents a link between the immune and endocrine system, which may form the gateway to the understanding of the initiation of thyroid autoimmune reactions and the thyroid autoimmune phenomena seen in iodine deficiency.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With regard to the involvement of the immune system in iodine-deficient
goitrogenesis, we (8, 9, 10) previously described an accumulation of
monocyte-derived cells such as macrophages (M
) and dendritic cells
(DC) in human endemic goitre and in the thyroid glands of rats and mice
a few weeks after starting an iodine-deficient diet. Because DC are
antigen-presenting cells (APC) par excellence and the only stimulators
of naive T lymphocytes, these early DC/M accumulations were proposed
as inducers of the relatively mild thyroid autoimmune reactions, which
often later develop in endemic goitre patients (11, 12) and
iodine-deficient rats (8). Indeed, the first local sign of an
autoimmune thyroiditis process is, in all animal models of spontaneous
disease studied so far, the accumulation and cluster formation of DC
and M in the thyroid gland (10, 13, 14). This accumulation occurs
prior to thyroid autoantibody formation and prior to the influx of
large numbers of T and B cells. Relevant to this report is that the
diabetic-prone Biobreeding (BB-DP) rat, one of the animal models of
spontaneous thyroid autoimmune disease, shows the formation of a small
goitre at the time of the early intrathyroid accumulation of DC (15).
The question thus arises whether there is a relationship between goitre
formation and DC accumulation.
There is a growing body of evidence that locally accumulated DC and
M are not only acting as APC and effector cells in host defence, but
also as cells involved in morphogenesis (woundhealing and matrix
repair; 16) and, more relevant to this report, in the regulation of
growth and function of various endocrine cells (17), i.e. of
anterior pituitary cells, ovarian granulosa cells, theca cells, luteal
cells, and testicular Leydig cells. M- and DC-derived cytokines,
interleukin-1 (IL-1), IL-6 and tumor necrosis factor (TNF), are thought
to play a role in such immuno-endocrine regulatory interactions
(17).
This notion has led to the aim of this study, i.e. the
investigation of the regulatory effects of rat splenic DC (excellent
APC) on particularly the growth, but also the function (thyroid hormone
secretion) of rat thyrocytes in a thyroid follicle culture system.
Splenic DC and thyroid follicles were isolated and enriched from
Wistar rats, and subsequently cocultured and their interactions studied
by time lapse cinematography. The tritiated thymidine
(3H-TdR) and bromodeoxyuridine (BrdU) incorporation as well
as the T3 release of thyrocytes exposed to DC under basal
conditions and under TSH stimulation was measured. Furthermore, the
effect of neutralizing antibodies to IL-1ß, IL-6, TNF-, and to
major histocompatibility complex-class I (MHC I), MHC II, intercellular
adhesion molecule-1 (ICAM-1) and lymphocyte function-associated
antigen-1 (LFA-1) was studied in this DC-thyroid follicle
coculture system.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Isolation and 24 h reconstitution culture of thyroid
follicles
Rats (females and males, 8–22 weeks of age) were asphyxiated
with carbon dioxide. Thyroid follicles were isolated as previously
described (15). Thyroid glands were excised and carefully trimmed of
fat and connective tissue. Subsequently, thyroid lobes were washed
(4 C) in calcium- and magnesium-free HBSS (Gibco BRL Life
Technologies, Breda, The Netherlands). The HBSS was removed and
replaced with dispersion solution, containing the following substances:
HBSS supplemented with 475 U/ml collagenase (type II; Sigma Chemical
Co., Axel, The Netherlands) and 0.1 mg/ml DNase (type I, grade II;
Boehringer Mannheim BV, Almere, The Netherlands). Thyroid lobes were
minced and incubated for 30 min (37 C, 5% CO2,
water-saturated atmosphere). The tissue was then further mechanically
disrupted by gentle pipetting. Thyroid fragments were left to sediment
during 1 min at 1 x g and released follicles present
in the supernatant were harvested and kept at 4 C. Fresh dispersion
solution was added to the remaining undigested tissue for 10 min and
the above described procedure was repeated. After several reincubation
steps the dissociation of thyroid tissue was complete.
The obtained suspension of thyroid follicles was washed twice with ice-cold HBSS containing 0.1 mg/ml DNase and centrifuged for 2 min at 50 x g. The pellet consisted of intact and partly fragmented follicles. Single follicular and endothelial cells, fibroblasts, and cell debris remaining in the supernatant were discarded. The isolated follicles were resuspended and cultured in Ham’s F-12 medium (Gibco), supplemented with a 5-hormone and peptide mixture containing 10 µg/ml insulin (Sigma), 10 nM hydrocortisone (Sigma), 5 µg/ml transferrin (Sigma), 10 ng/ml glycyl-L-histidyl-L-lysine-acetate (Sigma), 10 ng/ml somatostatin (Sigma) and 1% (vol/vol) heat-inactivated (30 min, 56 C) FCS (Integro BV, Zaandam, The Netherlands). The following antibiotics were also added: 100 U/ml penicillin (Seromed, Biochrom, Berlin, Germany), 0.1 mg/ml streptomycin (Seromed), and 2.5 µg/ml amphotericin (Boehringer).
At 24 h after initial isolation (37 C, 5% CO2, water-saturated atmosphere), the partly fragmented follicles became spheric-like structures again and contained 30–50 thyrocytes per follicle as judged by using an inverted microscope and hematoxylin-eosin staining, respectively.
Determination of thyrocyte proliferation in reconstituted thyroid
follicles stimulated for 24 h
The reconstituted follicles were washed and seeded in 24-wells
plates (Costar Europe Ltd., Badhoevedorp, The Netherlands) at an
averaged density of 1000 follicles/well (proliferation in thyroid
follicles, see below) or 2000 follicles/well (thyroid hormone secretion
by thyroid follicles, see below). Bovine-TSH (Sigma; concentrations
ranging from 1 · 10-1 to 5 · 102
mU/ml) and 0.1 µM KI (Merck, Darmstadt, Germany) were
added to the follicles. At this stage, splenic DC were also added at
various ratios to the thyroid follicles (see below, Coculture of
splenic dendritic cells and thyroid follicles). The follicles were
cultured for another 24 h. Basal controls consisted of cultures in
the absence of TSH.
The proliferation assays were designed to examine the entry of cells into the S-phase of the cell cycle. Two methods were used.
Tritiated thymidine uptake. Thyroid follicles were 3H-TdR labeled during the last 5 h of the 24 h exposure to TSH.
In brief, the follicles were washed and put on fresh Ham’s F12 medium (plus supplements and different concentrations TSH) in 96-wells plates (Costar). Subsequently, 1 µCi/well methyl-3H-TdR (Amersham International, Buckinghamshire, UK) was added. After the mentioned labeling period at 37 C in a 5% CO2 water-saturated incubator, the follicles were harvested on filter paper and radioactivity was counted in a liquid scintillation analyzer (LKB Betaplate, Wallac, Turku, Finland).
In some experiments, follicles were (a) cocultured with 5 µg/ml aphidicolin (inhibitor DNA polymerase, 18; Sigma) during or (b) 2,000 rad X-irradiated before TSH exposure. These treatments were applied to prevent replication of DNA, and thus to ascertain that measured radioactivity in the samples without these treatments reflect the S-phase of the cell cycle. In other words; exclusion of erroneous positive measurements, as discussed by Maurer (19). Each individual experiment was done in duplicate.
5-Bromo-2'-deoxyuridine labeling. The follicles were incubated at 37 C in a 5% CO2 water-saturated incubator in the presence of 10 µM BrdU (BrdU-kit I; Boehringer) during the last 5 h before the 24 h of TSH stimulation ended. Thereafter, the samples were thoroughly washed in 0.01 M PBS (Merck), cytocentrifuged onto glass slides, air-dried, and fixed in precooled 70% (vol/vol) ethanol (Merck) during exactly 30 min at -20 C.
The BrdU incorporation was immunocytochemically visualized according to the instructions of the manufacturer (see above) with slight modifications: 1) incubation with anti-BrdU antibody, diluted 1:10 in the manufacturer’s incubation buffer (containing nucleases that allow access to BrdU without denaturation of DNA), during 40 min at 37 C; 2) washing steps (3x) in PBS; 3) incubation with antimouse-immunoglobulins antibody, conjugated with fluorescein isothio-cyanate, diluted 1:10 in PBS containing 0.1% (wt/vol) BSA (Sigma), during 45 min at 37 C; 4) washing steps (3x) in PBS. The cytospin preparations were mounted in a solution containing 90% (vol/vol) glycerol (Sigma) and 2.3% (wt/vol) 1,4-diazobicyclo-(2,2,2)-octane (Sigma) to delay fading of the fluorescent dye during examination using an ordinary fluorescence microscope (Zeiss, Weesp, The Netherlands). The percentage BrdU-positive (BrdU+) follicles was evaluated by counting at least 500 follicles/sample. For negative controls, the primary antibody incubation step consisted of buffer solution without anti-BrdU antibody.
This thyroid follicle system also enabled us to determine the thyroid hormone secretion into the culture fluid. After 24 h of exposure to TSH, the culture supernatants were centrifuged (500 x g, 5 min), carefully aspirated (devoid of thyrocytes) and stored at -20 C until T3 levels were determined. The T3 contents in these supernatants were measured by a conventional RIA (Dr. T. J. Visser, Dr. R. Docter, Mr. H. van Toor, Department of Internal Medicine, University Hospital Dijkzigt, Rotterdam, The Netherlands). T3 standards were diluted in Ham’s F12 medium plus supplements. Each individual supernatant sample was assayed in duplicate.
Isolation of splenic dendritic cells and mixed leucocyte
reaction
Splenic DC from Wistar rats were enriched according to the
method of Knight et al. (20), with slight modifications.
Briefly, the spleen was minced and digested for 1 h at 37 C in
RPMI 1640 (Gibco) containing: 125 U/ml collagenase (type III; Gibco)
and 0.1 mg/ml DNase. The remaining tissue was teased through a (105
µm) filter and the erythrocytes were removed by lysis. Finally, the
cells were washed and cultured in RPMI 1640 supplemented with 10% FCS
and antibiotics. After an overnight culture period in culture flasks
(Costar; 37 C, 5% CO2 incubator), the DC were isolated
from the nonadherent cells by using a 14.5% (wt/vol) Nycodenz (Nycomed
Pharma As, Oslo, Norway) density gradient (800 x g for
20 min). Low density cells were collected from the interphase,
thoroughly washed in PBS/0.5% BSA and put on RPMI 1640 with 10% FCS
and antibiotics. This cell fraction demonstrated in 70–95% of cells a
dendritic morphology, a strong MHC II expression, and absence of acid
phosphatase activity.
For the mixed leucocyte reaction (MLR), splenic T lymphocytes were
isolated from the (high density cells) pellet fraction after 14.5%
Nycodenz density gradient centrifugation. T lymphocytes were separated
from B lymphocytes and some residual M by using a nylon wool column
(Polyscience Ltd., Eppelheim, Germany) during 1 h in 5%
CO2 incubator. These responder T (80–90%
CD3+) cells were used in the MLR by adding stimulator
(2,000 rad irradiated) DC at various ratios (fixed number of 150,000 T
cells/well) in a 96-wells plate, and were subsequently cultured for 5
days in RPMI 1640 containing 50 mM HEPES buffer (Gibco),
110 µg/ml Na-pyruvate (Merck), 0.5
(vol/vol) ß-mercapto-ethanol
(Merck) and antibiotics. T cell proliferation was measured via
3H-TdR incorporation (0.5 µCi/well during the last
16 h of total culture period).
To study the role of MHC and adhesion molecules in this MLR,
neutralizing antibodies against MHC I (rat RT1A; 1:1000; Serotec Ltd.,
Oxford, UK) and MHC II (rat RT1B; 1:400; Serotec), and against LFA-1
(rat CD11a; 1:50; Serotec) and ICAM-1 (rat CD54; 1:50; Serotec) were
added on the first and second day of a total culture period of 3 days.
Each individual experiment was done in triplicate.
Coculture of splenic dendritic cells and thyroid follicles
Splenic DC (in hormone supplemented Ham’s F-12 medium) were
2,000 rad X-irradiated to prevent proliferation and cocultured with
thyroid follicles at various ratios (10:1, 1:1 and 1:10) for 24 h
(with and without addition of TSH 100 mU/ml). Before coculturing, the
thyroid follicle suspensions (24 h after initial isolation) were
checked for possible presence of residual DC. A cytochemical staining
method (against MHC II) was used for this purpose and demonstrated that
these suspension contained no DC.
After the period of coculture, the effect of splenic DC on thyroid follicle 3H-TdR (and BrdU) uptake and T3 release was measured (see above, Determination of thyrocyte proliferation in reconstituted thyroid follicles stimulated for 24 h). The viability of the thyroid follicles and DC remained unchanged when cultured alone or cocultured (data not shown). Control cultures consisted of thyroid follicles and splenic DC alone (with and without TSH 100 mU/ml).
In some experiments (n = 3), the specificity of the DC effect on thyrocyte proliferation and T3 release was tested by fixing DC in a 4% (vol/vol) formaldehyde-PBS solution (Merck) for 20 min at room temperature before adding these DC to the follicles. This procedure prevented the motility of (as verified by time lapse video recording, see below) and production of cytokine(s) by DC.
To establish whether cytokines (IL-1ß, IL-6 and TNF-) or adhesive
interactions (MHC I and II, and LFA-1/ICAM-1) might contribute to
the effect of DC on the examined thyrocyte growth, the following
antibodies were added to our coculture system (DC:follicle ratio of
1:1) for 24 h: antihuman II-1ß (1:5000; Glaxo, Geneva,
Switzerland), antirat IL-6 (1:100; kindly provided by Dr. J. Gauldie,
Department of Pathology, McMaster University, Hamilton, Canada),
antihuman TNF- (1:1000; a kind gift from Dr. W. A. Buurman,
Department of Surgery, University of Limburg, Maastricht, The
Netherlands), antirat MHC I and II (see above, Isolation of
splenic dendritic cells and mixed leucocyte reaction), antirat
LFA-1 and ICAM-1 (see above). Each individual experiment was done in
duplicate.
Time lapse cinematography
To investigate DC motility and cell-cell interactions between DC
and thyroid follicles under the various above described coculture
conditions, time lapse cinematographic studies were performed. Again,
DC and thyroid follicles (ratio 1:10), with or without the various
anticytokines or antiadhesions, were cocultured and recorded during 45
min in a microincubator (37 C, 5% CO2) on an inverted
microscope (Axiovert, Zeiss) attached to a video camera (Sony Co.,
Japan) and a time lapse video recorder (Panasonic, Matsushita Electric
Industrial Co., Osaka, Japan). The video recording interval was set at
0.18 sec and the tape running speed at 2.599 mm/sec. After recording,
the images were analyzed using a computer (Acorn Computers Ltd.,
Cambridge, UK).
Statistical analysis
The results are presented as means ± SEM and
statistical analysis was performed using the two-tailed and paired
Students’ t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and C), 100 mU/ml being the most
optimal dosage for measuring both T3 release and
3H-TdR incorporation in our assay system. The specificity
of the 3H-TdR uptake with regard to true DNA proliferation
was demonstrated with aphidicolin and X-irradiation treatment. Both
treatments reduced >80% (n = 5) of the total measured
incorporated radioactivity (spontaneous and TSH-stimulated). This
reduction could not be ascribed to cytotoxicity of aphidicolin or
X-irradiation: none of the follicles showed a change in viability
(trypan blue exclusion test) or an altered morphology, and moreover the
T3 release remained unchanged (data not shown). In
accordance with the 3H-TdR uptake, our assay system showed
a clearly measurable BrdU labeling of the thyroid follicles with TSH at
100 and 500 mU/ml after 24 h of culture (Fig. 1B). Under basal
(non-TSH-stimulated conditions) < 5 BrdU+ nuclei per
follicle were always found, whereas exposure to TSH often resulted in
10–20 BrdU+ nuclei per follicle.
|
On the basis of the above-described findings, we decided to use a culture period of 24 h and a dosage of 100 mU/ml TSH for stimulation, and to consider the 3H-TdR incorporation as a reliable parameter of thyrocyte proliferation.
In a second set of experiments, we verified the accessory capability of
the splenic DC population. As expected, the Wistar rat splenic DC
(Nycodenz-enriched low density splenic cell fraction) were potent
stimulators of syngeneic T lymphocytes in MLR (Fig. 2). Already at a ratio of 1:40 (DC:T
cell) high incorporation values of around 70,000 cpm were found. In
contrast, resident peritoneal M or unfractionated splenic cells were
hardly capable of stimulating syngeneic T cells (values of at best
2,000–3,000 cpm were reached at a ratio 1:1), illustrating the superb
stimulatory capability of splenic DC.
|
). In the case of 3H-TdR
uptake values between 25–75% DC-induced inhibition were found
depending on the ratio DC:follicle and the conditions; the strongest
inhibition was found under basal conditions and at a ratio of 1:10
(follicle:DC), viz. 25 ± 7% of control (= thyroid follicle
culture in the absence of DCs, Fig. 3). The decreased BrdU labeling
largely confirmed the 3H-TdR uptake data, however, values
only (probably because of the semiquantitative nature of this method)
reached statistical significance under non-TSH-stimulated conditions
and at a ratio of 1:10 (follicle:DC, 71 ± 6% of control, Fig. 3). With regard to the T3 secretion, splenic DC induced a
subtle but consistent inhibition (i.e. 75 ± 7% of
control) on TSH-stimulated thyroid follicles at a ratio of 1:1 (Fig. 3). The T3 release in cocultures not stimulated with TSH
was not affected by the presence of splenic DC.
|
). They had more or less rounded or
elongated shapes and the vast majority were constantly extending and
withdrawing long cytoplasmic veils approaching and touching the thyroid
follicles. The dendritic (or veiled) cells often formed small cell
clusters with each other (containing 4–8 cells, so-called homotypic
clustering), and some of these DC clusters also made contact with the
thyroid follicles. The contacts of DC with the thyroid follicles varied
from short (mere touching at several spots of the thyroid follicle) to
longlasting and firm. During these firm contacts, the actively moving
DC were able to tilt a whole thyroid follicle. When formaldehyde-fixed
DC were used in this coculture system, their moving and active behavior
were completely lost as well as their homotypic clustering and
contacting with the thyroid follicles. This treatment also resulted in
a complete abrogation (data not shown; n = 3) of the inhibitory
effects of splenic DC on both the 3H-TdR uptake (basal and
TSH stimulated) and the T3 release (TSH stimulated).
|
, and
ICAM-1, whereas thyroid follicles were positive for ICAM-1 and LFA-1
(data not shown), we added anti-MHC I, anti-MHC II, anti-LFA-1, and
anti-ICAM-1 antibodies to the coculture system.
Before these experiments, the effect of these antibodies and their
effective dilutions were studied on homotypic clustering of splenic DC
and APC capability in syngeneic MLR. Effective dilutions for
interfering with DC clustering appeared to be 1:50 for anti-LFA-1
and anti-ICAM-1, 1:1000 for anti-MHC I, and 1:400 for anti-MHC II. The
homotypic clustering was delayed for several hours (2–3 h) by the
addition of anti-LFA-1 and anti-ICAM-1; normally homotypic
clustering starts at 30 min to 1 h. The anti-MHC I and anti-MHC II
intensified the homotypic clustering of DC and very large clusters of
over 50 cells (1.5 x 105 DC/well) were detected after
30 min in culture. With regard to the accessory capability of the
splenic DC, the antiadhesion and the anti-MHC antibodies clearly
reduced this function of these cells in syngeneic T cell MLR (Fig. 2
).
Despite their clear effects on the immune functions of the splenic DC,
the antibodies were not able, using the above indicated dilutions, to
interfere with the DC-induced inhibitory effects on 3H-TdR
uptake of thyroid follicles (Fig. 5, both
basal and TSH-stimulated conditions). In the absence of the splenic DC
these antibodies had also no effect on the 3H-TdR uptake of
thyrocytes (Fig. 5). By using time lapse cinematography it appeared
that the antibodies were effective in our follicle-DC coculture system
because homotypic DC clustering was affected in the system; the
interactions of splenic DC with thyroid follicles were, however,
similar in the presence or absence of anti-MHC I, anti-MHC II,
anti-LFA-1, and anti-ICAM-1.
|
were involved in
the antiproliferative effect of splenic DC on thyroid follicles,
experiments with neutralizing anticytokine antibodies were performed.
In earlier published experiments, we had established the effective
dilutions of these antibodies to neutralize the bioactivity of their
target cytokines in the appropriate bioassays (21, 22).
Anti-IL-1ß antibodies (1:5000) were able to abrogate the DC-induced
inhibition of 3H-TdR uptake in thyroid follicles, both
under basal and TSH-stimulated conditions (Fig. 6). This shows a key role of IL-1ß in
the thyrocyte antiproliferative effect of splenic DC. The
3H-TdR uptake of thyroid follicles in the absence of DC was
not affected by the anti-IL-1ß antibodies, demonstrating that the
cytokine was not involved in thyrocyte proliferation under these
conditions of the assay system.
|
).
This suggests an antiproliferative effect of thyrocyte-derived IL-6
under basal conditions (note: DC are absent from these reconstituted
follicles). The anti-IL-6 antibodies were also able to completely
abrogate the DC-induced inhibition of thyrocyte 3H-TdR
uptake under non-TSH-stimulated conditions (Fig. 6), hence an
involvement of this cytokine in DC-thyrocyte interactions is most
likely.
Anti-TNF- antibodies (1:1000) had a dampening effect
(P < 0.05) on TSH-stimulated 3H-TdR uptake
of thyroid follicles (Fig. 6), indicating that this cytokine
(thyrocyte-derived?) stimulates thyrocyte proliferation. No
neutralizing effect of anti-TNF- could be found on the thyrocyte
growth inhibiting effect of added splenic DC (Fig. 6).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in the pancreas, a
slight but consistent (25%) increase of islet and duct cell growth in
interferon-
transgenic mice, and Gaytan et al. (24)
observed raised testosterone levels in rats after DC/M depletion of
the testis. Whether an in vivo blockade of the intrathyroid
accumulation of DC (e.g. via the usage of the antibodies ED
7 and ED 8; Ref. 25) leads to an even more enhanced proliferation of
thyrocytes in iodine-deficient Wistar rat goitre or in BB-DP rat goitre
needs to be investigated.
Our study also shows that the cytokine IL-1ß is a key controlling
signal in the in vitro DC-down-regulated thyrocyte growth.
IL-6 also contributed to the immuno-endocrine interaction, but only
under non-TSH-stimulated conditions. TNF-, if any, had a growth
promoting effect in our culture system. Though epidermal DC (the
Langerhans cells) have been described as major sources of IL-1ß and
IL-6 (26), it is the general concept that cells belonging to the DC
series are poor producers of these cytokines. In our experience (lung
DC; 27) and that of others (28, 29), DC do not or hardly produce
IL-1ß and IL-6. Nevertheless, DC-derived IL-1ß signals play a major
role in DC-induced T cell proliferation as shown in cytokine
neutralization experiments (27) and apparently also in DC-thyrocyte
regulation (see this report).
Effects of IL-1ß and IL-6 on thyrocyte growth and function have amply
been described. In the majority of the reported experiments the
cytokines inhibit differentiated functions such as thyroid peroxidase
expression, thyroglobulin release, iodide uptake, and T3
secretion; reports on stimulation, however, also exist (reviewed in
Refs. 30, 31). The effect of these cytokines on thyrocyte
proliferation is also conjectural. Upon injection, IL-1 and IL-1ß
stimulate thyroid growth in rats (32), and IL-1 promotes the growth
of a thyroid carcinoma cell line via Ca2+/calmodulin
pathways (33). In contrast, the group of Gärtner found no effect
of IL-1ß on thyrocyte proliferation using a porcine thyroid follicle
culture system (34). IL-6 was described as not affecting basal cell
proliferation of FRTL-5 cells, it stimulated DNA-synthesis in a
dose-dependent manner in the presence of TSH (35). This growth
promoting effect of IL-6 is, however, controversial and Zeki et
al. (36) have reported very little effect on FRTL-5 cell
proliferation. Our data give evidence for a growth inhibiting effect of
DC-derived IL-1ß and IL-6. Obviously, culture conditions (the state
of the thyrocytes and the DC), local cytokine concentrations (31), and
possibly other cofactors provided in the close DC-thyrocyte
interactions may play a role as well. With regard to the
growth-inhibiting effect of DC-derived IL-1ß, it is also worthy to
note that recent reports show that this cytokine induces Fas expression
on thyrocytes, subsequently leading to the apoptosis of the thyrocytes
(37). Whether DC are able to induce Fas expression on rat thyrocytes in
our culture system needs further investigation.
Our cinematographic study showed that the DC were very active, moving
around and touching the thyroid follicles, while constantly extending
and withdrawing their veils. In some cases the DC made firm contacts
with the follicles and were able to tilt the follicles. It is possible
that these close contacts allow an efficient and special local cytokine
signalling between the DC and the thyrocytes, explaining the
above-discussed IL-1ß (and IL-6) induced inhibiting action on
thyrocyte proliferation. A specific triggering of the adhesion
molecules (receptors in itself) may also play a role in the DC-induced
thyrocyte growth inhibitory action; indeed, DC constitutively express
many adhesion molecules (38), whereas thyrocytes acquire such molecules
on their surface after cytokine/TSH stimulation (39). However,
antibodies to MHC I and II and to ICAM-1 and LFA-1 had no effect on
the contacting behavior and the growth controlling action of DC on
thyrocytes in coculture. These antibodies did show an effect on the
physical and functional interactions between DC and T lymphocytes. This
argues against at least a role of MHC molecules and LFA-1/ICAM-1 in
the functional contacts between DC and thyrocytes. It is obvious that
other adhesion molecules need to be tested in future experiments.
What could be the molecular signals attracting the DC to the thyroid? In the nonobese diabetic mouse, DC infiltration can be triggered by inducing thyrocyte necrosis via iodine-intoxication (10). In the thyroid of the BB-DP rat, the infiltration of DC occurs spontaneously at around 10 weeks of age, in the absence of early microscopic thyrocyte necrosis (13, 40, 41). At that time, isolated BB-DP thyrocytes produce in vitro more T4 and IL-6 as compared to age-matched Wistar thyrocytes (15), and these molecules may be instrumental in the local accumulation of DC because thyroid hormones and IL-6 are in vitro active in the monocyte to DC transition (42). Monocyte chemoattractant protein-1 (MCP-1) and granulocyte-macrophage colony-stimulating factor (GM-CSF) are known regulators for the attraction (43) and maturation (42) of monocyte-derived DC. However, MCP-1 is not expressed in the thyroid of 10-week-old BB-DP rats but becomes expressed later when lymphocytic infiltrates are present (15). IL-1 has been described as inducing MCP-1 and GM-CSF expression in cultured thyrocytes (44, 45). Hence, a scenario can be envisioned where DC accumulated in the thyroid by yet unknown signals (IL-6?) provide IL-1 stimuli to the neighboring thyrocytes, upon which these endocrine cells are not only regulated in their proliferation but also start to secrete cytokines (MCP-1, GM-CSF) that further stimulate the accumulation of DC.
In conclusion, the present study shows that classical APC, such as the DC, inhibit the in vitro growth (and hormonal output) of thyrocytes and that IL-1ß and to a lesser extent IL-6 are involved. Further studies are needed to establish the in vivo effects of DC on thyroid growth itself and a possible involvement of Fas molecules. Unveiling the effect of intrathyroidal DC on the growth of neighboring thyrocytes will probably lead to a better understanding of the initial interactions between thyrocytes and immune cells during the development of iodine-deficient goitre and thyroid autoimmune disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 2, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-Iodolactones
decrease epidermal growth factor-induced proliferation and
inositol-1,4,5-trishosphate generation in porcine thyroid follicles—a
possible mechanism of growth inhibition by iodide. Eur J Endocrinol 132:735–743 and
IL-1ß by psoriatic epidermal cells ex vivo. Eur J Immunol 25:1624–1630[Medline]
are selectively expressed in Langerhans cells
but are differentially regulated in culture. J Exp Med 176:1221–1226This article has been cited by other articles:
 |
|
 |
|
  H. A. Drexhage Are There More than Antibodies to the Thyroid-Stimulating Hormone Receptor that Meet the Eye in Graves' Disease? Endocrinology, January 1, 2006; 147(1): 9 - 12. [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Klein and H.-C. Wang Characterization of a novel set of resident intrathyroidal bone marrow-derived hematopoietic cells: potential for immune-endocrine interactions in thyroid homeostasis J. Exp. Biol., January 1, 2004; 207(1): 55 - 65. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. A. Delemarre, P. G. Hoogeveen, M. de Haan-Meulman, P. J. Simons, and H. A. Drexhage Homotypic cluster formation of dendritic cells, a close correlate of their state of maturation. Defects in the biobreeding diabetes-prone rat J. Leukoc. Biol., March 1, 2001; 69(3): 373 - 380. [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 |
