| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
, Interferon-
, and Transforming Growth Factor-ß on Adipogenesis and Expression of Thyrotropin Receptor in Human Orbital Preadipocyte Fibroblasts1
Division of Endocrinology, Metabolism and Nutrition, Mayo Clinic/Foundation, Rochester, Minnesota 55905
Address all correspondence and requests for reprints to: Rebecca S. Bahn, M.D., Mayo Clinic, Division of Endocrinology, 200 First Street, Southwest, Rochester, Minnesota 55905. E-mail: bahn.rebecca{at}mayo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TNF-), interferon-
(IFN-), or transforming growth factor-ß. We found that each
of these cytokines inhibits TSH-dependent cAMP production and TSHr gene
expression, and that TNF- and IFN- also inhibit morphological
adipocyte differentiation. When cytokines were added after
differentiation, the inhibition was less pronounced. Our results
suggest that TNF-, IFN-, and transforming growth factor-ß may
act within the orbit in GO to modulate expression of the putative
orbital autoantigen, TSHr. In addition, the former two cytokines may
play a role in determining the extent to which the volume of the
orbital adipose tissue increases in this condition. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Studies have shown that various inflammatory cytokines, including tumor
necrosis factor (TNF)-, interferon (IFN)-, and transforming
growth factor (TGF)-ß, are present in the orbit in GO. These and
other cytokines have been shown to increase glycosaminoglycan
production by orbital fibroblasts and to stimulate the expression by
these cells of various immunomodulatory proteins important in
pathogenesis of the disease (11, 12, 13, 14). The current studies
were designed to determine whether TNF-, IFN-, or TGF-ß might
also impact adipogenesis or TSHr expression in orbital preadipocyte
fibroblasts, thus suggesting another role for these cytokines in
GO.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Orbital adipose/connective tissue explants were obtained from patients undergoing orbital decompression surgery for severe GO (n = 4). All patients had been treated previously for Graves’ hyperthyroidism with 131I and were euthyroid on thyroid hormone replacement. Control normal orbital tissues were obtained during enucleation for corneal transplantation purpose from cadaveric donors with no history of GO or Graves’ disease (n = 4). GO or normal tissue samples were minced and placed directly in plastic culture dishes, allowing preadipocyte fibroblasts to proliferate as described previously (15). Cells were propagated in medium 199 containing 20% FBS (HyClone Laboratories, Inc., Logan, UT), penicillin (100 U/mL), and gentamicin (20 µg/mL) in a humidified 5% CO2 incubator at 37 C and maintained in 80-mm2 flasks with medium 199 containing 10% FBS and antibiotics.
To initiate adipocyte differentiation, orbital cells were grown to confluence in 6-well plates. Differentiation was carried out as reported previously (10); cultures were changed to serum-free DMEM/F12 (1:1, Sigma, St. Louis, MO) supplemented with biotin (33 µmol/L), pantothenic acid (17 µmol/L), transferrin (10 µg/mL), triiodothyronine (0.2 nmol/L), insulin (1 µmol/L), carbaprostacyclin (cPGI2; 0.2 µM; Calbiochem, La Jolla, CA), and (for the first 4 days only) dexamethasone (1 µmol/L) and isobutylmethylxanthine (IBMX; 0.1 mmol/L). The differentiation protocol was continued for 10–12 days, during which time the media were replaced every 3–4 days. Control fibroblasts, derived from the same patients’ orbital tissues, were cultured similarly, except for the omission in the medium of cPGI2, dexamethasone, and IBMX (16).
To evaluate the effect of cytokines during adipocyte differentiation,
we exposed cultures to recombinant human (rh) TNF- (Roche Molecular Biochemicals, Indianapolis, IN; 1 ng/mL),
rhIFN- (Roche Molecular Biochemicals; 1 ng/mL), or
human (h) TGF-ß (Collaborative Biomedical Products, Bedford, MA; 1
ng/mL) for the entire 10–12 day differentiation period. In other
experiments, we wished to determine the effect of treatment with these
cytokines on cells after differentiation. For these studies, cytokines
(10 ng/mL) were added to the culture media for only the last 24 h
of the differentiation period.
On days 1, 4, 7, and 10 of differentiation, cells were examined by phase-contrast microscopy, using an Axiovert 35 light microscope (Carl Zeiss, Thornwood, NY) equipped with a Contax 167 MT camera. Photographs were taken on Ektachrome Tungsten 64 colored film (Eastman Kodak, Rochester, NY).
cAMP measurement
Orbital cells in 6-well plates were preincubated in medium containing IBMX (1 mmol/L; 1 mL/well) for 2 h at 37 C. Before stimulation with rhTSH, both GO and normal orbital cells (whether subjected to differentiation or to the control protocol) produced basal quantities of cAMP that were generally below the sensitivity of the assay (typically less than 0.045 pmol/mL for the acetylated procedure). After the addition of rhTSH (3 x 10-7 mol/L; Genzyme Diagnostics, San Carlos, CA) to duplicate wells, incubation was continued for 2 h, until terminated with the addition of hydrochloric acid (0.1 mol/L; 1 mL/well) for 10 min. The culture media were subjected to centrifugation (600 x g), and cAMP production was measured using an acetylated procedure of a commercially available kit [cAMP (low pH) Immunoassay, R&D Systems, Minneapolis, MN]. Each raw data point represents the mean of duplicate determinations. For analyses, each value was normalized to 100%, representing the maximal cAMP production by differentiated cells in that particular experiment.
Ribonuclease (RNase) protection assay
After experiments in vitro, cells were pelleted and stored frozen at -70 C. Total RNA was isolated directly from frozen specimens using the Totally RNA Kit (Ambion, Inc., Austin, TX). Positive-control RNA was prepared, in the same manner, from cultured Chinese hamster ovary cells that had been transfected with plasmid containing the hTSHr (JPO9 line) or from a negative control counterpart (JPO2 line).
The antisense RNA probe for TSHr was transcribed from a 320-bp PCR product with a T7 phage promoter at its 3-prime end, in the presence of T7 RNA polymerase (10 U) and [32P]uridine 5'-triphosphate (50 µCi) for labeling (8). The resulting high-specific-activity probe encompassed nucleotides 576–873 (exons 6–9) of the hTSHr complementary DNA sequence, as reported by Nagayama (17), and was designed to detect both the 2.4-kb intact TSHr (protecting a product of 298 nucleotides) and the 1.3-kb variant form (protecting a product of 217 nucleotides). The antisense RNA probe for human glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was generated from pTRI-GAPDH human antisense control template (Ambion, Inc.). This probe was designed to protect a 154-nucleotide fragment of GAPDH messenger RNA (mRNA).
Total RNA (80 µg) was combined with 300,000 cpm TSHr probe and 3,000 cpm GAPDH probe in hybridization buffer, denatured at 95 C, and hybridized at 45 C for 16 h. Nonhybridized total RNA and probe were digested for 1 h at 37 C with RNase A (0.175 U) and RNase T1 (25 U; RNase Protection Kit, Roche Molecular Biochemicals). Samples were subsequently digested with proteinase K (50 µg) in the presence of 0.5% SDS and extracted with phenol/chloroform/isoamyl alcohol. The resulting ethanol-precipitated protected fragments were resuspended in loading buffer and resolved on a denaturing polyacrylamide gel (5% acrylamide/8 mol/L urea).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The production of cAMP, in response to rhTSH stimulation, was
measured in cells after culture for 10–12 days in
adipocyte-differentiation media. Paired cultures of cells, derived from
the same GO patient or normal individual, were grown in control media
for the same period of time. Before stimulation with rhTSH, both GO and
normal orbital cells (whether subjected to differentiation or to the
control protocol) produced quantities of cAMP that were generally below
the sensitivity of the assay (typically less than 0.045 pmol/mL for the
acetylated procedure). After stimulation with rhTSH (3 x
10-7 mol/L), cAMP
production was found uniformly to be measurable and greater in
differentiated orbital cultures than in undifferentiated control
cultures of cells from the same individual (Table 1
). Differentiated orbital cultures
obtained from normal individuals generally produced more cAMP in
response to rhTSH stimulation (mean, 45 pmol/mL; range, 11.3–68.7
pmol/mL) than did differentiated cultures of GO cells (mean, 9 pmol/mL;
range, 0.9–20.3 pmol/mL). Similarly, control cultures from normal
individuals showed generally higher absolute levels of rhTSH-dependent
cAMP production (mean, 13.7 pmol/mL; range, 1.8–30.9 pmol/mL) than did
GO cells cultured in the same fashion (mean, 1.0 pmol/mL; range,
undetectable-2.6 pmol/mL). These results, showing quantitative
differences in cAMP production between GO and normal cultures, were
similar to our previously reported findings, which also demonstrated
the variability of this response between cells derived from different
individuals (10).
|
Treatment of orbital fibroblasts with rhTNF-, rhIFN-, or
hTGF-ß during the entire 10-day differentiation period resulted in
profound inhibition of TSH-dependent cAMP production (mean: 99%, 95%,
and 95% inhibition, respectively; Fig. 1). Levels of cAMP in these cultures were
even lower than those measured in the control cultures that were
maintained for the same period of time in medium lacking several of the
components necessary for complete adipocyte differentiation. The same
was the case after treatment of normal orbital cells with rhTNF- or
rhIFN- during differentiation (mean: 96% and 97% inhibition,
respectively; Fig. 2). However, hTGF-ß
treatment of normal cells resulted in less inhibition of TSH-dependent
cAMP production (mean, 47% inhibition; Fig. 2) than was observed in
the GO cultures. The absolute levels of TSH-dependent cAMP production,
in cells exposed to cytokines during differentiation, were generally
greater in normal than in GO cultures (Table 1
).
|
|
Treatment of cultures of GO orbital fibroblasts with rhIFN-,
during the final 24 h of the 10-day differentiation protocol,
resulted in inhibition of TSH-dependent cAMP production in GO (mean,
80% inhibition; Fig. 3) and normal
(mean, 55% inhibition; Fig. 4) orbital
preadipocyte fibroblasts. Though treatment of cultures with rhTNF-
or hTGF-ß, during this time, also resulted in inhibition in GO (mean:
30% and 15% inhibition, respectively; Fig. 3) and normal (mean: 24%
and 25% inhibition, respectively; Fig. 4) orbital cultures, the degree
of inhibition in each was less than that seen after rhIFN-
treatment. The absolute levels of TSH-dependent cAMP production were
generally greater in normal than in GO cells exposed to cytokines after
differentiation (Table 1).
|
|
(10 ng/mL) for the final 24 h of the 10-day
differentiation period. We found an inhibition of TSHr mRNA expression
in these cultures, compared with cultures not treated with cytokine
(Fig. 5).
|
We examined the morphology of differentiated, control, and
cytokine-treated cultures using phase-contrast microscopy. Within the
first 24 h of culture, most of the GO or normal cells exposed to
the differentiation protocol lost their elongated fibroblast-like
appearance and became rounded. By day 4 of differentiation, some of
these cells formed droplets that seemed to contain lipid. These
droplets were clearly evident by days 7–10 of differentiation (Fig. 6A). In contrast, the majority of GO or
normal cells exposed to control conditions still maintained a
fibroblast-like appearance at 24 h in culture. By days 7–10, some
of the control cells were somewhat rounded in appearance, but no
droplets could be detected (Fig. 6B). Cultures treated with cytokines
showed morphology that depended on the particular cytokine present in
the culture media during differentiation. Normal or GO cells exposed to
either rhTNF- (Fig. 6C) or rhIFN-, throughout the entire 10-day
protocol, showed changes that were similar to those observed in cells
grown in control media (i.e. rounding of cells, with no
evidence of droplet formation). In contrast, hTGF-ß-treated cells
seemed to differentiate normally and formed droplets similarly to those
of cells grown under the differentiation conditions (Fig. 6D).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adipocyte precursor cells can be isolated from the stromal-vascular fraction of porcine and human adipose/connective tissues (26, 27). These cells are thought to be a subpopulation of fibroblasts (so-called: preadipocyte fibroblasts) that can undergo adipocyte differentiation when cultured under appropriate conditions (28). Sorisky and colleagues (16) demonstrated that 5–10% of cells contained in cultures of human orbital fibroblasts are capable of in vitro adipogenesis. Haraguchi and colleagues (29) reported that differentiation of rat preadipocytes, under similar conditions, is closely related to increased expression of TSHr in these cells. Our recent studies support and link both of these observations. We demonstrated increased expression of TSHr in human orbital fibroblast cultures exposed to conditions, that resulted in a subpopulation of these cells acquiring the morphologic features of adipocytes (10).
Several cytokines (including IFN-, TNF-, IL-2, IL-4, IL-5, and
IL-10) have been shown to be present within the retroocular tissues of
patients with severe GO (12, 30). These inflammatory
cytokines are likely secreted by mononuclear cells that infiltrate the
orbit in GO. Studies in vitro have shown that IFN-
treatment of orbital fibroblasts increases glycosaminoglycan production
and that both IFN- and TNF- enhance expression in orbital
fibroblasts of human leukocyte antigen-DR, intercellular adhesion
molecule-1 and heat shock protein-72, immunomodulatory molecules
(3). Because such cytokine effects occurring within the
orbit would likely result in the histopathology characteristic of GO,
these cytokines are thought to play a role in the pathogenesis of GO
(31).
In the current study, we showed profound inhibition of TSH-dependent
cAMP production after treatment of GO orbital cells with TNF-,
IFN-, or TGF-ß during the 10-day differentiation period. This
inhibition was not attributable to any differences in the number of
cells present in the cytokine-treated cultures, compared with the
untreated cultures after differentiation. In addition, we found
morphological differentiation of the cells to be partially inhibited
when either IFN- or TNF- was present in cultures during the
entire differentiation process. In contrast, TGF-ß treatment did not
significantly impact morphological differentiation. Thus, it seems that
IFN- or TNF- likely inhibits adipocyte differentiation in these
cells. As a result, the accompanying functional expression of TSHr does
not develop. In contrast, TGF-ß seems to allow adipocyte
differentiation to progress normally while inhibiting functional TSHr
expression. Whether TSHr gene expression itself or only TSH-dependent
cAMP production is affected by this cytokine is unclear at present.
These findings suggest that TSHr expression may not be essential for
adipogenesis but may be a characteristic feature of newly
differentiated adipocytes. Alternately, it is possible that TSHr is
also expressed at a point in the differentiation process before the
final stages of adipogenesis. When cytokines (particularly IFN-) are
added after morphological differentiation seems to be complete
(i.e. during the final 24 h of culture), both
TSH-dependent cAMP production and TSHr gene expression decrease. Thus,
cytokines may also act to partially reverse in vitro
adipogenesis in these cells.
Our results are somewhat similar to those reported for rat epididymal
preadipocyte cultures in which exposure to TNF-, IFN-, or TGF-ß
during differentiation inhibited morphological differentiation, TSHr
gene expression, and TSH-dependent cAMP production (32).
These cytokines, and especially TNF-, have also been shown in other
systems to decrease lipid and carbohydrate metabolism, inhibit
adipocyte differentiation, and induce dedifferentiation of the fully
differentiated adipocyte phenotype (33, 34, 35, 36, 37). The
mechanisms by which TNF- affects adipocyte differentiation may
involve down-regulation of PPAR, a positive regulator of adipocyte
differentiation (38, 39), modulation of prostaglandin
synthesis (40), and/or activation of the
mitogen-activated protein kinase signaling pathway
(41). In addition, because this cytokine has been
implicated in apoptosis of 3T3-L1 cells (42) and human
adipose cells (43), it is possible that apoptosis plays a
role in the inhibition of adipocyte differentiation by TNF-.
In summary, we report that TNF-, IFN-, and TGF-ß decrease the
expression of functional TSHr in human orbital preadipocyte fibroblasts
exposed to adipogenesis-stimulating culture conditions. In addition,
TNF- and IFN- may also inhibit adipogenesis in these cells. Thus,
these cytokines would be expected to decrease expression of the
putative orbital autoantigen in GO and to counter the expansion of
tissues within the orbit. Both of these effects would likely aid in the
resolution of the orbital disease process. These findings would seem to
run counter to our earlier studies showing effects of TNF-, IFN-,
and TGF-ß on orbital fibroblasts thought to be important in the
progression of GO, including stimulation of glycosaminoglycan
production and expression of immunomodulatory proteins
(31). This apparent contradiction can be reconciled by the
fact that cytokines are pleotropic effectors with multiple properties
(44). The dominant effect of a particular cytokine at a
certain point in the course of a disease is influenced by the presence
and relative magnitude of many other disease-related factors.
We hypothesize that TSHr expression in the orbit is stimulated in Graves’ disease by some (as yet, undefined) factor(s). The increased expression of this autoantigen results in infiltration of the orbit by TSHr-activated, circulating T cells. Many different cytokines are produced and released within the orbit by these T cells, as well as by orbital fibroblasts and macrophages. A complex interplay between the pleotropic effects that these factors exert on the orbital tissues likely determines the severity and course of the disease process in GO.
|
Footnotes |
|---|
Received July 24, 2000.
Revised September 30, 2000.
Accepted October 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in cultured fibroblasts
derived from patients with Graves’ ophthalmopathy and pretibial
dermopathy. J Clin Endocrinol Metab. 73:307–313.[Abstract]
and interleukin-1ß inhibit
adipoconversion in cultured rodent preadipocytes. J Cell Physiol. 151:300–309.[CrossRef][Medline]
gene
expression contributes to the antiadipogenic effects of tumor necrosis
factor-. Mol Endocrinol. 10:1457–1466.[Abstract]
- mediated inhibition and reversal of
adipocyte differentiation is accompanied by suppressed expression of
PPAR without effects on Pref-1 expression. Endocrinology. 138:2776–2783. inversely regulates prostaglandin D2
and prostaglandin E2 production in murine
macrophages. J Biol Chem. 272:31065–31072.
disrupts mitotic clonal expansion and regulation of retinoblastoma
proteins p130 and p107 during 3T3–L1 adipocyte differentiation. Biochem Biophys Res Commun. 247:373–378.[CrossRef][Medline]
induces apoptosis of human adipose cells. Diabetes. 46:1939–1944.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  I. J Bujalska, O. M Durrani, J. Abbott, C. U Onyimba, P. Khosla, A. H Moosavi, T. T Q Reuser, P. M Stewart, J. W Tomlinson, E. A Walker, et al. Characterisation of 11{beta}-hydroxysteroid dehydrogenase 1 in human orbital adipose tissue: a comparison with subcutaneous and omental fat J. Endocrinol., February 1, 2007; 192(2): 279 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Cawood, P. Moriarty, C. O'Farrelly, and D. O'Shea Smoking and Thyroid-Associated Ophthalmopathy: A Novel Explanation of the Biological Link J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 59 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T J Cawood, P Moriarty, C O'Farrelly, and D O'Shea The effects of tumour necrosis factor-{alpha} and interleukin1 on an in vitro model of thyroid-associated ophthalmopathy; contrasting effects on adipogenesis Eur. J. Endocrinol., September 1, 2006; 155(3): 395 - 403. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Schaffler, U. Muller-Ladner, J. Scholmerich, and C. Buchler Role of Adipose Tissue as an Inflammatory Organ in Human Diseases Endocr. Rev., August 1, 2006; 27(5): 449 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Prabhakar, R. S. Bahn, and T. J. Smith Current Perspective on the Pathogenesis of Graves' Disease and Ophthalmopathy Endocr. Rev., December 1, 2003; 24(6): 802 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Bahn Pathophysiology of Graves' Ophthalmopathy: The Cycle of Disease J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 1939 - 1946. [Full Text] [PDF]
|
|
|
|
|
|
R. W. Valyasevi, D. A. Harteneck, C. M. Dutton, and R. S. Bahn Stimulation of Adipogenesis, Peroxisome Proliferator-Activated Receptor-{gamma} (PPAR{gamma}), and Thyrotropin Receptor by PPAR{gamma} Agonist in Human Orbital Preadipocyte Fibroblasts J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2352 - 2358. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
